chore(master): release 1.4.2 (#48 )

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
test: add 129 tests for critical untested code (core, CoSTEER, factor_coder, model_coder, qlib pipeline)
2026-07-29 16:37:43 +00:00 · 2026-05-03 11:02:32 +02:00 · 2026-05-03 10:59:24 +02:00 · 2026-05-03 10:19:59 +02:00 · 2026-05-03 09:39:56 +02:00 · 2026-05-03 09:37:00 +02:00
122 changed files with 8815 additions and 1148 deletions
@@ -14,7 +14,7 @@ jobs:
  security:
    runs-on: ubuntu-latest
    steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6

      - name: Run Bandit (Security Scan)
        uses: PyCQA/bandit-action@v1
@@ -25,9 +25,9 @@ jobs:
  test:
    runs-on: ubuntu-latest
    steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6

-      - uses: actions/setup-python@v5
+      - uses: actions/setup-python@v6
        with:
          python-version: "3.10"
          cache: "pip"
@@ -36,11 +36,11 @@ jobs:
    steps:
      # Checkout the repository to the GitHub Actions runner
      - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6

      # Execute Codacy Analysis CLI and generate a SARIF output with the security issues identified during the analysis
      - name: Run Codacy Analysis CLI
-        uses: codacy/codacy-analysis-cli-action@d840f886c4bd4edc059706d09c6a1586111c540b
+        uses: codacy/codacy-analysis-cli-action@562ee3e92b8e92df8b67e0a5ff8aa8e261919c08
        env:
          JAVA_TOOL_OPTIONS: "-Dfile.encoding=UTF-8"
        with:
@@ -46,7 +46,7 @@ jobs:
    name: Validate Commit Messages
    runs-on: ubuntu-latest
    steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
        with:
          fetch-depth: 0

@@ -25,10 +25,10 @@ jobs:

    steps:
    - name: Checkout repository
-      uses: actions/checkout@v4
+      uses: actions/checkout@v6

    - name: Set up Python
-      uses: actions/setup-python@v5
+      uses: actions/setup-python@v6
      with:
        python-version: "3.10"

@@ -64,7 +64,7 @@ jobs:

    - name: Upload docs artifact
      if: github.ref == 'refs/heads/main'
-      uses: actions/upload-pages-artifact@v3
+      uses: actions/upload-pages-artifact@v5
      with:
        path: docs/_build/html

@@ -16,10 +16,10 @@ jobs:

    steps:
    - name: Checkout repository
-      uses: actions/checkout@v4
+      uses: actions/checkout@v6

    - name: Set up Python
-      uses: actions/setup-python@v5
+      uses: actions/setup-python@v6
      with:
        python-version: "3.10"

@@ -12,7 +12,8 @@ jobs:
  release-please:
    runs-on: ubuntu-latest
    steps:
-      - uses: googleapis/release-please-action@v4
+      - uses: googleapis/release-please-action@v5
        with:
-          release-type: python
          token: ${{ secrets.GITHUB_TOKEN }}
+          config-file: release-please-config.json
+          manifest-file: .release-please-manifest.json
@@ -19,9 +19,9 @@ jobs:
        python-version: ["3.10", "3.11"]

    steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6

-      - uses: actions/setup-python@v5
+      - uses: actions/setup-python@v6
        with:
          python-version: ${{ matrix.python-version }}
          cache: "pip"
@@ -49,9 +49,9 @@ jobs:
    name: Dependency Audit
    runs-on: ubuntu-latest
    steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6

-      - uses: actions/setup-python@v5
+      - uses: actions/setup-python@v6
        with:
          python-version: "3.10"
          cache: "pip"
@@ -19,10 +19,10 @@ jobs:

    steps:
    - name: Checkout repository
-      uses: actions/checkout@v4
+      uses: actions/checkout@v6

    - name: Set up Python
-      uses: actions/setup-python@v5
+      uses: actions/setup-python@v6
      with:
        python-version: "3.10"

@@ -0,0 +1,3 @@
+{
+  ".": "1.4.2"
+}
@@ -1,5 +1,172 @@
 # Changelog

+## [1.4.2](https://github.com/TPTBusiness/Predix/compare/v1.4.1...v1.4.2) (2026-05-03)
+
+
+### Bug Fixes
+
+* add missing sys import and fix undefined acc_rate in factor eval ([c45f990](https://github.com/TPTBusiness/Predix/commit/c45f9908ee321400f0a19c57f1482e4cd1394a50))
+
+## [1.4.1](https://github.com/TPTBusiness/Predix/compare/v1.4.0...v1.4.1) (2026-05-03)
+
+
+### Bug Fixes
+
+* 15 bug fixes across orchestrator, runner, backtest, and infrastructure ([163687d](https://github.com/TPTBusiness/Predix/commit/163687d7e1c278a085d7052a3f958a3edb501e77))
+* also catch ValueError in mean_variance for dimension mismatch ([ed73b72](https://github.com/TPTBusiness/Predix/commit/ed73b7253f7dc6459ee30dd81a1ce1194e46e9af))
+* close log file handle, fix FTMO equity double-count, remove bare except ([76219a5](https://github.com/TPTBusiness/Predix/commit/76219a53efddaafc2b8bd48a0f76c1d4325e6ea5))
+* correct project root paths and subprocess handling in parallel runner and CLI ([9735e3a](https://github.com/TPTBusiness/Predix/commit/9735e3a4d8f01e7b16fb9b185a002396a915cea4))
+* filter NaN in max(), remove redundant ternary, handle non-finite vbt results ([f89fbb3](https://github.com/TPTBusiness/Predix/commit/f89fbb3421faf6ccdc8e68a911fd9db2c166120f))
+* fix type annotation, remove unused parameter, improve import_class errors ([8b6ab73](https://github.com/TPTBusiness/Predix/commit/8b6ab735c05629bf6b76ddc2fd8b15617600cad7))
+* resolve dead code, shell injection risk, mutable defaults, and other bugs ([afff262](https://github.com/TPTBusiness/Predix/commit/afff26287f7c4df7ddfde4e816d280fe845e11eb))
+* resolve unbound variable, logger shadowing, withdraw_loop edge case, and other bugs in main scripts ([748cf9b](https://github.com/TPTBusiness/Predix/commit/748cf9b214a3e8447f1289fc4cf1e92ad6cc2f1a))
+
+## [1.4.0](https://github.com/TPTBusiness/Predix/compare/v1.3.11...v1.4.0) (2026-05-01)
+
+
+### Features
+
+* **optimizer:** add max_positions parameter to Optuna search space ([fdb4be3](https://github.com/TPTBusiness/Predix/commit/fdb4be3b3ebd93325e7821f4251148424184a40d))
+
+## [1.3.11](https://github.com/TPTBusiness/Predix/compare/v1.3.10...v1.3.11) (2026-05-01)
+
+
+### Bug Fixes
+
+* **ci:** lazy import logger in predix.py and cli.py to avoid ImportError in test env ([60763e8](https://github.com/TPTBusiness/Predix/commit/60763e8eae34f41865ba8e5e65bdfde13b564b4b))
+
+## [1.3.10](https://github.com/TPTBusiness/Predix/compare/v1.3.9...v1.3.10) (2026-05-01)
+
+
+### Bug Fixes
+
+* **security:** replace remaining assert statements with proper error handling ([928533d](https://github.com/TPTBusiness/Predix/commit/928533d9a81bd5062f07458fbf94d3c7fe347775))
+
+## [1.3.9](https://github.com/TPTBusiness/Predix/compare/v1.3.8...v1.3.9) (2026-05-01)
+
+
+### Bug Fixes
+
+* **security:** resolve path-injection, B701, B101, B112 Bandit alerts ([20b89a0](https://github.com/TPTBusiness/Predix/commit/20b89a061843b39836e975f158404e8e2d4627cd))
+
+## [1.3.8](https://github.com/TPTBusiness/Predix/compare/v1.3.7...v1.3.8) (2026-04-30)
+
+
+### Bug Fixes
+
+* **deps:** relax aiohttp constraint to &gt;=3.13.4 for litellm compatibility ([34ab192](https://github.com/TPTBusiness/Predix/commit/34ab1923a887089eb36e5cbad6cb8df16f0333ca))
+* **qlib:** correct indentation in except blocks in quant_proposal and factor_runner ([8143451](https://github.com/TPTBusiness/Predix/commit/8143451e8c0ead01c4d86d19669268c7bfb15fac))
+* **security:** replace eval() with ast.literal_eval in finetune validator (B307) ([0508caf](https://github.com/TPTBusiness/Predix/commit/0508caf9140d210b823fefefa28ee535ec85a0ae))
+* **security:** replace shell=True subprocess calls with list args in env.py (B602) ([2012d5a](https://github.com/TPTBusiness/Predix/commit/2012d5ae4e77cc2f1ab9a48beaaac5a74695d083))
+* **security:** resolve path-injection and add nosec for safe temp paths (B108, py/path-injection) ([6727480](https://github.com/TPTBusiness/Predix/commit/67274803bd1d14e5d1df9a063f46b2edb8501a2b))
+
+## [1.3.7](https://github.com/TPTBusiness/Predix/compare/v1.3.6...v1.3.7) (2026-04-30)
+
+
+### Bug Fixes
+
+* **security:** nosec for B608/B701 false positives in UI and template code ([5eb5d7e](https://github.com/TPTBusiness/Predix/commit/5eb5d7e8fdbe90e0dced83fef4e09f5a33e96b2b))
+* **security:** replace eval() with ast.literal_eval and add request timeouts (B307, B113) ([3301ada](https://github.com/TPTBusiness/Predix/commit/3301ada697ca7d3afa1a188d2a76a87ae98b4529))
+* **security:** replace shell=True subprocess calls with list args (B602) ([13c08f4](https://github.com/TPTBusiness/Predix/commit/13c08f4ce6813eb7c314087921ec8c0f40074bd7))
+
+## [1.3.6](https://github.com/TPTBusiness/Predix/compare/v1.3.5...v1.3.6) (2026-04-30)
+
+
+### Bug Fixes
+
+* **security:** real fix for B110 (logging in factor_proposal.py [#746](https://github.com/TPTBusiness/Predix/issues/746)) ([16624e0](https://github.com/TPTBusiness/Predix/commit/16624e0bd966ae4d24c4a3eb42bbc31c11da3136))
+* **security:** real fix for B110 (logging in factor_runner.py [#744](https://github.com/TPTBusiness/Predix/issues/744)) ([88cf0fb](https://github.com/TPTBusiness/Predix/commit/88cf0fb8828b11c97f2f3ae2881a4900b020c6f0))
+* **security:** real fix for B110 (logging in quant_proposal.py [#741](https://github.com/TPTBusiness/Predix/issues/741)) ([7cf2a64](https://github.com/TPTBusiness/Predix/commit/7cf2a644f553b054bd4b0607ea51e5372e68d90a))
+* **security:** real fix for B110 (logging in quant_proposal.py [#741](https://github.com/TPTBusiness/Predix/issues/741)) ([ef985f8](https://github.com/TPTBusiness/Predix/commit/ef985f86035d8dca707c60137e6508349a0c4ae6))
+* **security:** real fix for B404/B603 (sys.executable in factor_runner.py [#745](https://github.com/TPTBusiness/Predix/issues/745)) ([819655a](https://github.com/TPTBusiness/Predix/commit/819655aaa3efa76596d60501d0e8ca365df3e5e2))
+* **security:** revert broken read_pickle encoding arg in kaggle template (B301) ([3574907](https://github.com/TPTBusiness/Predix/commit/35749073c91e69f63ddaad61dae3f2b799327e63))
+* **security:** validate SQL identifiers in _add_column_if_not_exists (B608) ([e10dfa2](https://github.com/TPTBusiness/Predix/commit/e10dfa2576038e911f83595d3b466c261bc0cd54))
+* **security:** whitelist-validate metric column in get_top_factors (B608) ([e50519f](https://github.com/TPTBusiness/Predix/commit/e50519fe066e68aec2f19b83df4f643c3c22053d))
+
+## [1.3.5](https://github.com/TPTBusiness/Predix/compare/v1.3.4...v1.3.5) (2026-04-27)
+
+
+### Bug Fixes
+
+* **auto-fixer:** add five new factor code fixes for groupby/apply errors ([449c8fd](https://github.com/TPTBusiness/Predix/commit/449c8fd70a327e604dcca122e4a134f0cca918e4))
+* **auto-fixer:** add four new factor code fixes for common runtime errors ([40484f6](https://github.com/TPTBusiness/Predix/commit/40484f6d300425da481f1edd325da4acbc06ec7d))
+* **auto-fixer:** add groupby([level=N,'date']) SyntaxError fix ([ca77c00](https://github.com/TPTBusiness/Predix/commit/ca77c005bea4abdd8854c1de2b0e8d03b7742161))
+* **auto-fixer:** disable _fix_min_periods for intraday data ([77b0740](https://github.com/TPTBusiness/Predix/commit/77b0740f059349df7e769a378af728aa33b2070e))
+* **auto-fixer:** fix chained groupby(level=N).groupby('date') pattern ([7d5fe32](https://github.com/TPTBusiness/Predix/commit/7d5fe32b31a19ce8b04bd8f5a430720fdb748f7a))
+* **auto-fixer:** fix df.loc[instrument] DateParseError on MultiIndex frames ([b7860ea](https://github.com/TPTBusiness/Predix/commit/b7860eafc0ad26384947ce0510ecf4e9f3425807))
+* **auto-fixer:** fix df['instrument'] KeyError on MultiIndex frames ([aad6bd1](https://github.com/TPTBusiness/Predix/commit/aad6bd1c7c720b3d486e0cf248337f32394773b1))
+* **auto-fixer:** fix two assignment-target bugs in instrument column fixers ([421eedf](https://github.com/TPTBusiness/Predix/commit/421eedffed4b883c24397dc5581c019a3985277f))
+* **auto-fixer:** preserve date dimension in groupby(['instrument','date']) fix ([b58fdd8](https://github.com/TPTBusiness/Predix/commit/b58fdd8be43720b5d4363e0f8de9a01591d4d2dc))
+* **auto-fixer:** remove ddof from rolling() args, not only from std()/var() ([b0fc328](https://github.com/TPTBusiness/Predix/commit/b0fc328d0d4a041c65d8eeb32cb3f2bb86568406))
+* **auto-fixer:** strip spurious .reset_index() after .transform() calls ([8708aae](https://github.com/TPTBusiness/Predix/commit/8708aae6e08728cda1875c775a76dc92e43576f3))
+* **loop:** prevent step_idx advance on unhandled exceptions + fix consecutive assistant messages ([5ec4ad1](https://github.com/TPTBusiness/Predix/commit/5ec4ad1b96b5b99ef42bea7bb828cb1ef709a688))
+
+## [1.3.4](https://github.com/TPTBusiness/Predix/compare/v1.3.3...v1.3.4) (2026-04-27)
+
+
+### Bug Fixes
+
+* **auto-fixer:** add five new factor code fixes for groupby/apply errors ([449c8fd](https://github.com/TPTBusiness/Predix/commit/449c8fd70a327e604dcca122e4a134f0cca918e4))
+* **auto-fixer:** add four new factor code fixes for common runtime errors ([40484f6](https://github.com/TPTBusiness/Predix/commit/40484f6d300425da481f1edd325da4acbc06ec7d))
+* **auto-fixer:** add groupby([level=N,'date']) SyntaxError fix ([ca77c00](https://github.com/TPTBusiness/Predix/commit/ca77c005bea4abdd8854c1de2b0e8d03b7742161))
+* **auto-fixer:** disable _fix_min_periods for intraday data ([77b0740](https://github.com/TPTBusiness/Predix/commit/77b0740f059349df7e769a378af728aa33b2070e))
+* **auto-fixer:** fix chained groupby(level=N).groupby('date') pattern ([7d5fe32](https://github.com/TPTBusiness/Predix/commit/7d5fe32b31a19ce8b04bd8f5a430720fdb748f7a))
+* **auto-fixer:** fix df.loc[instrument] DateParseError on MultiIndex frames ([b7860ea](https://github.com/TPTBusiness/Predix/commit/b7860eafc0ad26384947ce0510ecf4e9f3425807))
+* **auto-fixer:** fix df['instrument'] KeyError on MultiIndex frames ([aad6bd1](https://github.com/TPTBusiness/Predix/commit/aad6bd1c7c720b3d486e0cf248337f32394773b1))
+* **auto-fixer:** preserve date dimension in groupby(['instrument','date']) fix ([b58fdd8](https://github.com/TPTBusiness/Predix/commit/b58fdd8be43720b5d4363e0f8de9a01591d4d2dc))
+* **auto-fixer:** remove ddof from rolling() args, not only from std()/var() ([b0fc328](https://github.com/TPTBusiness/Predix/commit/b0fc328d0d4a041c65d8eeb32cb3f2bb86568406))
+* **backtest:** replace broken MC permutation test with binomial win-rate test ([c38d894](https://github.com/TPTBusiness/Predix/commit/c38d89478f586825bfca5715a96ca70ccd8791a3))
+* **factors:** detect and correct look-ahead bias in daily-constant factors ([eb490a4](https://github.com/TPTBusiness/Predix/commit/eb490a461b66cbd815ae53ac5205115754712432))
+* **factors:** extend look-ahead rules to session factors and add intraday-factor guidance ([c24c100](https://github.com/TPTBusiness/Predix/commit/c24c100442d6487686c0578de0b32d240fcbf215))
+* **loop:** compress old experiment history in proposal prompt to reduce context size ([4bf90a9](https://github.com/TPTBusiness/Predix/commit/4bf90a905ba8b2aba2a818191c19998088cccaaf))
+* **loop:** prevent step_idx advance on unhandled exceptions + fix consecutive assistant messages ([5ec4ad1](https://github.com/TPTBusiness/Predix/commit/5ec4ad1b96b5b99ef42bea7bb828cb1ef709a688))
+
+## [1.3.3](https://github.com/TPTBusiness/Predix/compare/v1.3.2...v1.3.3) (2026-04-25)
+
+
+### Bug Fixes
+
+* **backtest:** replace broken MC permutation test with binomial win-rate test ([c38d894](https://github.com/TPTBusiness/Predix/commit/c38d89478f586825bfca5715a96ca70ccd8791a3))
+* **factors:** detect and correct look-ahead bias in daily-constant factors ([eb490a4](https://github.com/TPTBusiness/Predix/commit/eb490a461b66cbd815ae53ac5205115754712432))
+* **factors:** extend look-ahead rules to session factors and add intraday-factor guidance ([c24c100](https://github.com/TPTBusiness/Predix/commit/c24c100442d6487686c0578de0b32d240fcbf215))
+* **loop:** compress old experiment history in proposal prompt to reduce context size ([4bf90a9](https://github.com/TPTBusiness/Predix/commit/4bf90a905ba8b2aba2a818191c19998088cccaaf))
+* **strategies:** guard against None IC in acceptance check, disable slow wf_rolling ([2197f52](https://github.com/TPTBusiness/Predix/commit/2197f52150a50ef38d9e70991d7e48c8c30caec4))
+* **strategies:** handle None ic/sharpe/dd in rejected strategy log output ([ad2ad3a](https://github.com/TPTBusiness/Predix/commit/ad2ad3ab3360ea75ed3bbc90c12098b9c5cc0114))
+
+## [1.3.2](https://github.com/TPTBusiness/Predix/compare/v1.3.1...v1.3.2) (2026-04-23)
+
+
+### Bug Fixes
+
+* **strategies:** guard against None IC in acceptance check, disable slow wf_rolling ([2197f52](https://github.com/TPTBusiness/Predix/commit/2197f52150a50ef38d9e70991d7e48c8c30caec4))
+* **strategies:** handle None ic/sharpe/dd in rejected strategy log output ([ad2ad3a](https://github.com/TPTBusiness/Predix/commit/ad2ad3ab3360ea75ed3bbc90c12098b9c5cc0114))
+
+## [1.3.1](https://github.com/TPTBusiness/Predix/compare/v1.3.0...v1.3.1) (2026-04-21)
+
+
+### Bug Fixes
+
+* **deps:** bump python-dotenv to &gt;=1.2.2 (CVE symlink overwrite) ([126ae7d](https://github.com/TPTBusiness/Predix/commit/126ae7d5fb556b677d09d10221862a0d648d697a))
+
+## [1.3.0](https://github.com/TPTBusiness/Predix/compare/v1.2.2...v1.3.0) (2026-04-21)
+
+
+### Features
+
+* **backtest:** add rolling walk-forward validation and Monte Carlo trade permutation test ([637a94c](https://github.com/TPTBusiness/Predix/commit/637a94c1d987da763869f4f9b73372a3f37d873c))
+
+
+### Bug Fixes
+
+* **security:** resolve all 30 Bandit security alerts (B301, B614, B104) ([ce5983d](https://github.com/TPTBusiness/Predix/commit/ce5983d9d59c4c34341fb1ec749e44bbcfc4a1c4))
+
+## [1.2.2](https://github.com/TPTBusiness/Predix/compare/v1.2.1...v1.2.2) (2026-04-19)
+
+
+### Documentation
+
+* **claude:** auto-merge release-please PR after every push ([f500917](https://github.com/TPTBusiness/Predix/commit/f500917b699ee78dc676e84e01574d49bdc8e796))
+
 ## [2.2.0](https://github.com/TPTBusiness/Predix/compare/v2.1.0...v2.2.0) (2026-04-18)


@@ -13,11 +13,19 @@ import sys
 from pathlib import Path

 from dotenv import load_dotenv
+
 load_dotenv(Path(__file__).parent / ".env")

 import typer
 from rich.console import Console

+try:
+    from rdagent.utils.env import logger
+except ImportError:
+    import logging
+
+    logger = logging.getLogger(__name__)
+
 app = typer.Typer(help="Predix - AI Quantitative Trading Agent")
 console = Console()

@@ -107,7 +115,7 @@ def _ensure_kronos_factor_in_pool(con) -> None:
        color = "green" if abs(ic) > 0.01 else "yellow"
        con.print(
            f"  [bold {color}]Kronos Factor ready:[/bold {color}] IC={ic:.4f}, "
-            f"Hit-Rate={hit_rate:.1%} — added to strategy pool"
+            f"Hit-Rate={hit_rate:.1%} — added to strategy pool",
        )

    except Exception as e:
@@ -194,9 +202,9 @@ def quant(
        predix health - Check system health and configuration
    """
    import subprocess
+    import sys
    import threading
    import time
-    import sys

    # ---- Parallel Run Isolation ----
    # When run_id > 0, isolate all outputs (logs, results, workspace)
@@ -219,10 +227,9 @@ def quant(
        console.print(f"   [dim]Log: {log_file}[/dim]")
        console.print(f"   [dim]Results: results/runs/run{run_id}/[/dim]")
        console.print(f"   [dim]Workspace: {workspace_dir.name}/[/dim]")
-    else:
-        # Single run mode: default log file
-        if log_file is None:
-            log_file = "fin_quant.log"
+    # Single run mode: default log file
+    elif log_file is None:
+        log_file = "fin_quant.log"

    # ---- Log File Setup (daily-rotated) ----
    from datetime import datetime as _dt
@@ -230,10 +237,14 @@ def quant(
    _daily_dir = Path(__file__).parent / "logs" / _today
    _daily_dir.mkdir(parents=True, exist_ok=True)

+    _log_f = None
+    _orig_stdout = sys.stdout
+    _orig_stderr = sys.stderr
+
    if log_file.lower() != "none":
        log_path = _daily_dir / log_file
        # Open log file for appending (raw stdout/stderr capture)
-        log_f = open(log_path, "a", encoding="utf-8")
+        _log_f = open(log_path, "a", encoding="utf-8")

        # Redirect stdout and stderr to both console and log file
        class TeeWriter:
@@ -245,18 +256,18 @@ def quant(
                    try:
                        s.write(data)
                        s.flush()
-                    except:
+                    except Exception:
                        pass

            def flush(self):
                for s in self._streams:
                    try:
                        s.flush()
-                    except:
+                    except Exception:
                        pass

-        sys.stdout = TeeWriter(sys.__stdout__, log_f)
-        sys.stderr = TeeWriter(sys.__stderr__, log_f)
+        sys.stdout = TeeWriter(_orig_stdout, _log_f)
+        sys.stderr = TeeWriter(_orig_stderr, _log_f)

        console.print(f"\n[dim]📝 Logging to: logs/{_today}/{log_file}[/dim]")
    else:
@@ -269,7 +280,7 @@ def quant(
        if not api_key:
            console.print("\n[bold red]❌ OPENROUTER_API_KEY not set in .env[/bold red]")
            console.print("[yellow]Add your API key to .env:[/yellow]")
-            console.print('  OPENROUTER_API_KEY=sk-or-your-key-here')
+            console.print("  OPENROUTER_API_KEY=sk-or-your-key-here")
            raise typer.Exit(code=1)

        # Setup both API keys for load balancing
@@ -282,7 +293,7 @@ def quant(
            os.environ["LITELLM_PARALLEL_CALLS"] = "2"
            console.print(f"\n[bold blue]🌐 Using OpenRouter (2 API Keys):[/bold blue] [cyan]{os.environ['CHAT_MODEL']}[/cyan]")
            console.print(f"   [dim]Keys: {api_key[:15]}*** + {api_key_2[:15]}***[/dim]")
-            console.print(f"   [dim]Parallel: 2 concurrent requests[/dim]")
+            console.print("   [dim]Parallel: 2 concurrent requests[/dim]")
        else:
            os.environ["OPENAI_API_KEY"] = api_key
            console.print(f"\n[bold blue]🌐 Using OpenRouter:[/bold blue] [cyan]{os.environ['CHAT_MODEL']}[/cyan]")
@@ -300,7 +311,7 @@ def quant(
    # ---- Dashboards ----
    if dashboard:
        def start_web_dashboard():
-            console.print(f"\n[bold green]🚀 Web Dashboard: http://localhost:5000[/bold green]")
+            console.print("\n[bold green]🚀 Web Dashboard: http://localhost:5000[/bold green]")
            subprocess.run(
                ["python", "web/dashboard_api.py"],
                cwd=str(Path(__file__).parent),
@@ -325,7 +336,7 @@ def quant(
    from rdagent.app.qlib_rd_loop.quant import main as fin_quant
    from rdagent.log.daily_log import session as _daily_session

-    console.print(f"\n[bold cyan]📊 Starting EURUSD Trading Loop...[/bold cyan]\n")
+    console.print("\n[bold cyan]📊 Starting EURUSD Trading Loop...[/bold cyan]\n")

    _ctx = {"model": model}
    if run_id:
@@ -335,11 +346,17 @@ def quant(
    if step_n:
        _ctx["steps"] = step_n

-    with _daily_session("fin_quant", **_ctx):
-        fin_quant(
-            step_n=step_n,
-            loop_n=loop_n,
-        )
+    try:
+        with _daily_session("fin_quant", **_ctx):
+            fin_quant(
+                step_n=step_n,
+                loop_n=loop_n,
+            )
+    finally:
+        if _log_f is not None:
+            sys.stdout = _orig_stdout
+            sys.stderr = _orig_stderr
+            _log_f.close()


@app.command()
@@ -408,8 +425,8 @@ def evaluate(
        predix portfolio - Select a diversified portfolio of uncorrelated factors
        predix quant - Generate new factors via LLM trading loop
    """
-    from rich.panel import Panel
    from rdagent.log.daily_log import session as _daily_session
+    from rich.panel import Panel

    console.print(Panel(
        "[bold cyan]📊 Predix Factor Evaluator[/bold cyan]\n"
@@ -489,11 +506,12 @@ def top(
        predix portfolio - Select diversified portfolio from top factors
        predix build-strategies - Combine factors into trading strategies
    """
-    import json
    import glob as glob_module
+    import json
+
    import numpy as np
-    from rich.table import Table
    from rich.panel import Panel
+    from rich.table import Table

    factors_dir = Path(__file__).parent / "results" / "factors"
    if not factors_dir.exists():
@@ -510,6 +528,7 @@ def top(
            if data.get("status") == "success" and data.get("ic") is not None:
                results.append(data)
        except Exception:
+            logger.warning("Failed to load factor file %s", f, exc_info=True)
            continue

    if not results:
@@ -557,9 +576,11 @@ def top(

    console.print(table)

-    # Summary
-    valid_ic = [r.get("ic") for r in results if r.get("ic") is not None]
-    valid_sharpe = [r.get("sharpe") for r in results if r.get("sharpe") is not None]
+    # Summary — filter None, NaN, and non-numeric values
+    valid_ic = [v for v in (r.get("ic") for r in results)
+                if isinstance(v, (int, float)) and v is not None and not np.isnan(v)]
+    valid_sharpe = [v for v in (r.get("sharpe") for r in results)
+                    if isinstance(v, (int, float)) and v is not None and not np.isnan(v)]
    # Filter extreme outliers for average
    valid_sharpe_filtered = [s for s in valid_sharpe if abs(s or 0) < 1e6]

@@ -634,16 +655,16 @@ def portfolio(
        predix top - View top factors before portfolio selection
        predix build-strategies - Build strategies from selected factors
    """
-    import json
    import glob as glob_module
+    import json
+    import shutil
    import subprocess
    import tempfile
-    import shutil
-    import numpy as np
+
    import pandas as pd
-    from rich.table import Table
    from rich.panel import Panel
-    from rich.progress import Progress, SpinnerColumn, TextColumn, BarColumn, TaskProgressColumn, TimeElapsedColumn
+    from rich.progress import BarColumn, Progress, SpinnerColumn, TaskProgressColumn, TextColumn, TimeElapsedColumn
+    from rich.table import Table

    factors_dir = Path(__file__).parent / "results" / "factors"
    if not factors_dir.exists():
@@ -659,6 +680,7 @@ def portfolio(
            if data.get("status") == "success" and data.get("ic") is not None:
                results.append(data)
        except Exception:
+            logger.warning("Failed to load factor file %s", f, exc_info=True)
            continue

    if not results:
@@ -674,12 +696,12 @@ def portfolio(
    # 2. Evaluate candidates to get time-series values for correlation
    # We need to run the factor code to get the series of values.
    # We do this sequentially to avoid OOM.
-    
+
    # Locate data file
    data_file = Path(__file__).parent / "git_ignore_folder" / "factor_implementation_source_data" / "intraday_pv.h5"
    if not data_file.exists():
        data_file = Path(__file__).parent / "git_ignore_folder" / "factor_implementation_source_data_debug" / "intraday_pv.h5"
-    
+
    if not data_file.exists():
        console.print("[red]Source data file (intraday_pv.h5) not found.[/red]")
        return
@@ -696,11 +718,11 @@ def portfolio(
        console=console,
    ) as progress:
        task = progress.add_task(f"Computing values for {len(candidates)} factors...", total=len(candidates))
-        
+
        for cand in candidates:
            fname = cand.get("factor_name", "unknown")
            fcode = cand.get("factor_code", "")
-            
+
            if not fcode:
                errors.append((fname, "No code in JSON"))
                progress.advance(task)
@@ -716,10 +738,10 @@ def portfolio(
                    # If symlink fails, copy the file
                    import shutil
                    shutil.copy(str(data_file), str(tmp_path / "intraday_pv.h5"))
-                
+
                # Write code
                (tmp_path / "factor.py").write_text(fcode)
-                
+
                try:
                    # Run factor
                    result = subprocess.run(
@@ -727,16 +749,16 @@ def portfolio(
                        cwd=tmp_path,
                        capture_output=True,
                        text=True,
-                        timeout=120 # 2 min timeout per factor
+                        timeout=120, # 2 min timeout per factor
                    )
-                    
+
                    # Read result
                    res_file = tmp_path / "result.h5"
                    if res_file.exists():
                        df = pd.read_hdf(str(res_file), key="data")
                        # Get the series (first column)
                        series = df.iloc[:, 0]
-                        
+
                        # Count non-NaN values
                        non_nan = series.count()
                        if non_nan < 1000:
@@ -756,7 +778,7 @@ def portfolio(
                except Exception as e:
                    errors.append((fname, str(e)[:100]))
                    progress.update(task, description=f"{fname} ❌ (Error)")
-            
+
            progress.advance(task)

    # Show summary of errors
@@ -770,7 +792,7 @@ def portfolio(
    if len(factor_series) < 3:
        console.print("[red]Not enough valid factor series to build portfolio (need at least 3).[/red]")
        console.print("[yellow]Tip: Factors might be producing mostly NaN values or failing execution.[/yellow]")
-        
+
        # Fallback: Show top factors by IC without diversification
        console.print("\n[dim]Showing top factors by IC instead:[/dim]")
        table = Table(
@@ -788,50 +810,49 @@ def portfolio(
                str(i),
                cand.get("factor_name", "unknown")[:38],
                f"{cand.get('ic', 0):.6f}",
-                f"{cand.get('sharpe', 0):.4f}" if cand.get('sharpe') else "N/A",
+                f"{cand.get('sharpe', 0):.4f}" if cand.get("sharpe") else "N/A",
            )
-        
+
        console.print(table)
        return

    # 3. Build Correlation Matrix
    console.print(f"\n[dim]Building correlation matrix from {len(factor_series)} factors...[/dim]")
-    
+
    # Align indices and drop NaN
    combined = pd.DataFrame(factor_series).dropna()
-    
+
    if combined.empty or len(combined) < 100:
        console.print("[red]Not enough valid overlapping data to compute correlation.[/red]")
        console.print("[dim]This means the factors produce values at different times or have too many NaN values.[/dim]")
        return

    corr_matrix = combined.corr().fillna(0)
-    ic_map = {cand['factor_name']: cand.get('ic', 0) for cand in candidates}
+    ic_map = {cand["factor_name"]: cand.get("ic", 0) for cand in candidates}

    # 4. Greedy Selection
    selected = []
    remaining = list(corr_matrix.columns)
-    
+
    # Sort remaining by IC to prioritize high IC factors
    remaining.sort(key=lambda x: abs(ic_map.get(x, 0)), reverse=True)

    for factor in remaining:
        if len(selected) >= target:
            break
-        
+
        # If it's the first one, just take it
        if not selected:
            selected.append(factor)
            continue
-        
+
        # Check correlation with already selected
        # We want max(|corr|) < max_corr
        max_c = 0
        for sel in selected:
            c = abs(corr_matrix.loc[factor, sel])
-            if c > max_c:
-                max_c = c
-        
+            max_c = max(max_c, c)
+
        if max_c < max_corr:
            selected.append(factor)

@@ -849,23 +870,23 @@ def portfolio(

    for i, fname in enumerate(selected, 1):
        # Find original data for display
-        data = next((c for c in candidates if c['factor_name'] == fname), {})
-        ic = data.get('ic')
-        sharpe = data.get('sharpe')
-        
+        data = next((c for c in candidates if c["factor_name"] == fname), {})
+        ic = data.get("ic")
+        sharpe = data.get("sharpe")
+
        # Calculate max corr with other selected factors
        max_c_val = 0
        for s in selected:
            if s != fname:
                val = abs(corr_matrix.loc[fname, s])
-                if val > max_c_val: max_c_val = val
+                max_c_val = max(max_c_val, val)

        table.add_row(
            str(i),
            fname[:38],
            f"{ic:.6f}" if ic is not None else "N/A",
            f"{sharpe:.4f}" if sharpe is not None else "N/A",
-            f"{max_c_val:.4f}" if max_c_val > 0 else "-"
+            f"{max_c_val:.4f}" if max_c_val > 0 else "-",
        )

    console.print(table)
@@ -875,20 +896,20 @@ def portfolio(
        "selected_factors": selected,
        "max_correlation": max_corr,
        "pool_size": top,
-        "timestamp": pd.Timestamp.now().isoformat()
+        "timestamp": pd.Timestamp.now().isoformat(),
    }
-    
+
    out_dir = Path(__file__).parent / "results" / "portfolio"
    out_dir.mkdir(parents=True, exist_ok=True)
    out_file = out_dir / "selected_factors.json"
-    
+
    with open(out_file, "w") as f:
        json.dump(portfolio_data, f, indent=2)
-    
+
    console.print(Panel(
        f"[bold]Portfolio saved to results/portfolio/selected_factors.json[/bold]\n"
        f"Selected {len(selected)} unique factors from {top} candidates.",
-        border_style="green"
+        border_style="green",
    ))


@@ -934,13 +955,12 @@ def portfolio_simple(
        predix top - View top factors before portfolio selection
        predix build-strategies - Build strategies from selected factors
    """
-    import json
    import glob as glob_module
-    import re
-    import numpy as np
+    import json
+
    import pandas as pd
-    from rich.table import Table
    from rich.panel import Panel
+    from rich.table import Table

    factors_dir = Path(__file__).parent / "results" / "factors"
    if not factors_dir.exists():
@@ -956,6 +976,7 @@ def portfolio_simple(
            if data.get("status") == "success" and data.get("ic") is not None:
                results.append(data)
        except Exception:
+            logger.warning("Failed to load factor file %s", f, exc_info=True)
            continue

    if not results:
@@ -983,14 +1004,14 @@ def portfolio_simple(
    for cand in candidates:
        fname = cand.get("factor_name", "").lower()
        assigned = False
-        
+
        # Check each category's keywords
        for cat, keywords in categories.items():
            if any(kw in fname for kw in keywords):
                categorized[cat].append(cand)
                assigned = True
                break
-        
+
        if not assigned:
            categorized["other"].append(cand)

@@ -1001,7 +1022,7 @@ def portfolio_simple(
            best = categorized[cat][0]  # Already sorted by IC
            selected.append({
                "factor": best,
-                "category": cat.capitalize() if cat != "other" else "Other"
+                "category": cat.capitalize() if cat != "other" else "Other",
            })

    # 5. Display Results
@@ -1024,7 +1045,7 @@ def portfolio_simple(
            cand.get("factor_name", "unknown")[:38],
            cat,
            f"{cand.get('ic', 0):.6f}",
-            f"{cand.get('sharpe', 0):.4f}" if cand.get('sharpe') else "N/A",
+            f"{cand.get('sharpe', 0):.4f}" if cand.get("sharpe") else "N/A",
        )

    console.print(table)
@@ -1034,7 +1055,7 @@ def portfolio_simple(
        "selected_factors": [item["factor"]["factor_name"] for item in selected],
        "categories": {item["category"]: item["factor"]["factor_name"] for item in selected},
        "method": "simple_keyword_categorization",
-        "timestamp": str(pd.Timestamp.now().isoformat())
+        "timestamp": str(pd.Timestamp.now().isoformat()),
    }

    out_dir = Path(__file__).parent / "results" / "portfolio"
@@ -1047,7 +1068,7 @@ def portfolio_simple(
    console.print(Panel(
        f"[bold]Simple Portfolio saved to results/portfolio/portfolio_simple.json[/bold]\n"
        f"Selected {len(selected)} factors across {len([c for c in categorized if categorized[c]])} categories.",
-        border_style="green"
+        border_style="green",
    ))


@@ -1110,12 +1131,10 @@ def build_strategies(
        predix portfolio - Select diversified factors before combining
        predix top - View top factors before building strategies
    """
-    import pandas as pd
    import numpy as np
-    from rich.table import Table
-    from rich.panel import Panel
-
    from rdagent.scenarios.qlib.developer.strategy_builder import StrategyBuilder
+    from rich.panel import Panel
+    from rich.table import Table

    console.print(Panel(
        "[bold cyan]🏗️  Predix Strategy Builder[/bold cyan]\n"
@@ -1271,9 +1290,10 @@ def build_strategies_ai(
        predix quant - Generate new alpha factors via LLM trading loop
        predix evaluate - Evaluate factors before strategy building
    """
-    from rich.panel import Panel
    from pathlib import Path

+    from rich.panel import Panel
+
    console.print(Panel(
        "[bold cyan]🧠 StrategyCoSTEER - AI Strategy Builder[/bold cyan]\n"
        "Generating trading strategies from existing factors\n"
@@ -1299,7 +1319,7 @@ def build_strategies_ai(
    # Setup LLM environment (same as quant command)
    api_key = os.getenv("OPENROUTER_API_KEY") or os.getenv("OPENAI_API_KEY", "")
    api_key_2 = os.getenv("OPENROUTER_API_KEY_2", "")
-    
+
    if api_key and not api_key.startswith("sk-or-"):
        # OPENROUTER_API_KEY not set, try to use what we have
        api_key = os.getenv("OPENROUTER_API_KEY", api_key)
@@ -1326,8 +1346,8 @@ def build_strategies_ai(
        return

    # Load evaluated factors
-    import json
    import glob as glob_module
+    import json

    factors = []
    for f in glob_module.glob(str(factors_dir / "*.json")):
@@ -1337,6 +1357,7 @@ def build_strategies_ai(
            if data.get("status") == "success" and data.get("ic") is not None:
                factors.append(data)
        except Exception:
+            logger.warning("Failed to load factor file %s", f, exc_info=True)
            continue

    if len(factors) < 10:
@@ -1410,15 +1431,15 @@ def build_strategies_ai(

            for i, r in enumerate(results, 1):
                # Monthly return: use real backtest if available, else estimate
-                rb = r.get('real_backtest', {})
-                if isinstance(rb, dict) and rb.get('status') == 'success':
-                    monthly_pct = rb.get('monthly_return_pct', r.get('monthly_return_pct', 0))
-                    n_trades = rb.get('n_trades', '-')
-                    real_ic = rb.get('ic', 0)
+                rb = r.get("real_backtest", {})
+                if isinstance(rb, dict) and rb.get("status") == "success":
+                    monthly_pct = rb.get("monthly_return_pct", r.get("monthly_return_pct", 0))
+                    n_trades = rb.get("n_trades", "-")
+                    real_ic = rb.get("ic", 0)
                else:
-                    monthly_pct = r.get('monthly_return_pct', r.get('real_monthly_return', 0))
-                    n_trades = '-'
-                    real_ic = rb.get('ic', 0) if isinstance(rb, dict) else 0
+                    monthly_pct = r.get("monthly_return_pct", r.get("real_monthly_return", 0))
+                    n_trades = "-"
+                    real_ic = rb.get("ic", 0) if isinstance(rb, dict) else 0

                table.add_row(
                    str(i),
@@ -1511,7 +1532,7 @@ def status():
    # Process check
    result = subprocess.run(
        ["pgrep", "-f", "fin_quant"],
-        capture_output=True, text=True
+        capture_output=True, text=True,
    )
    if result.returncode == 0:
        console.print("[bold green]✅ Trading Loop: RUNNING[/bold green]")
@@ -1529,7 +1550,7 @@ def status():
        factors = c.fetchone()[0]
        conn.close()

-        console.print(f"\n📊 Results:")
+        console.print("\n📊 Results:")
        console.print(f"  Backtest runs: {runs}")
        console.print(f"  Factors: {factors}")

@@ -1552,6 +1573,7 @@ def _load_strategies():
            try:
                raw = json.loads(p.read_text())
            except Exception:
+                logger.warning("Failed to load strategy file %s", p, exc_info=True)
                continue
            if not isinstance(raw, dict):
                continue
@@ -1605,6 +1627,7 @@ def best(
        $ predix best -n 50 --export /tmp/top.json
    """
    import json
+
    from rich.table import Table

    items = _load_strategies()
@@ -1721,7 +1744,7 @@ def kronos_factor(
        console.print("Run data conversion first — see README Data Setup section.")
        raise typer.Exit(1)

-    console.print(f"[bold]Kronos Factor Generator[/bold]")
+    console.print("[bold]Kronos Factor Generator[/bold]")
    console.print(f"  Context: [cyan]{context}[/cyan] bars  |  Pred: [cyan]{pred}[/cyan] bars  |  Device: [cyan]{_device}[/cyan]")

    from rdagent.components.coder.kronos_adapter import build_kronos_factor
@@ -1804,7 +1827,7 @@ def kronos_eval(
        console.print(f"[red]ERROR: Data not found at {data_path}[/red]")
        raise typer.Exit(1)

-    console.print(f"[bold]Kronos Model Evaluator[/bold] (alongside LightGBM)")
+    console.print("[bold]Kronos Model Evaluator[/bold] (alongside LightGBM)")
    console.print(f"  Context: [cyan]{context}[/cyan] bars  |  Pred: [cyan]{pred}[/cyan] bars  |  Device: [cyan]{_device}[/cyan]")
    console.print("  Running evaluation...")

@@ -1819,12 +1842,12 @@ def kronos_eval(
        batch_size=batch_size,
    )

-    console.print(f"\n[bold]Kronos-mini Results[/bold]")
+    console.print("\n[bold]Kronos-mini Results[/bold]")
    console.print(f"  Predictions: [cyan]{metrics['n_predictions']}[/cyan]")
    console.print(f"  IC (mean):   [{'green' if metrics['IC_mean'] > 0.02 else 'yellow'}]{metrics['IC_mean']:.4f}[/]")
    console.print(f"  IC IR:       [{'green' if metrics['IC_IR'] > 0.5 else 'yellow'}]{metrics['IC_IR']:.4f}[/]  (>0.5 = strong signal)")
    console.print(f"  Hit Rate:    [{'green' if metrics['hit_rate'] > 0.52 else 'yellow'}]{metrics['hit_rate']:.2%}[/]  (>50% = directionally useful)")
-    console.print(f"\n[dim]Reference: LightGBM baseline IC typically 0.01–0.05 on 1-min EUR/USD[/dim]")
+    console.print("\n[dim]Reference: LightGBM baseline IC typically 0.01–0.05 on 1-min EUR/USD[/dim]")

    import json as _json
    out_dir = Path("results/kronos")
@@ -68,7 +68,7 @@ ignore_missing_imports = true
 module = "llama"

 [tool.pytest.ini_options]
-addopts = "-l -s --durations=0"
+addopts = "-l -s --durations=0 -m 'not slow'"
 log_cli = true
 log_cli_level = "info"
 log_date_format = "%Y-%m-%d %H:%M:%S"
@@ -21,11 +21,17 @@ load_dotenv(".env")

 import subprocess
 from importlib.resources import path as rpath
-from typing import Dict, Optional
+from typing import Annotated

 import typer
 from rich.console import Console
-from typing_extensions import Annotated
+
+try:
+    from rdagent.utils.env import logger
+except ImportError:
+    import logging
+
+    logger = logging.getLogger(__name__)

 from rdagent.app.data_science.loop import main as data_science
 from rdagent.app.finetune.llm.loop import main as llm_finetune
@@ -139,10 +145,10 @@ def ds_user_interact(port=19900):

@app.command(name="fin_factor")
 def fin_factor_cli(
-    path: Optional[str] = None,
-    step_n: Optional[int] = None,
-    loop_n: Optional[int] = None,
-    all_duration: Optional[str] = None,
+    path: str | None = None,
+    step_n: int | None = None,
+    loop_n: int | None = None,
+    all_duration: str | None = None,
    checkout: CheckoutOption = True,
 ):
    fin_factor(path=path, step_n=step_n, loop_n=loop_n, all_duration=all_duration, checkout=checkout)
@@ -150,10 +156,10 @@ def fin_factor_cli(

@app.command(name="fin_model")
 def fin_model_cli(
-    path: Optional[str] = None,
-    step_n: Optional[int] = None,
-    loop_n: Optional[int] = None,
-    all_duration: Optional[str] = None,
+    path: str | None = None,
+    step_n: int | None = None,
+    loop_n: int | None = None,
+    all_duration: str | None = None,
    checkout: CheckoutOption = True,
 ):
    fin_model(path=path, step_n=step_n, loop_n=loop_n, all_duration=all_duration, checkout=checkout)
@@ -161,10 +167,10 @@ def fin_model_cli(

@app.command(name="fin_quant")
 def fin_quant_cli(
-    path: Optional[str] = None,
-    step_n: Optional[int] = None,
-    loop_n: Optional[int] = None,
-    all_duration: Optional[str] = None,
+    path: str | None = None,
+    step_n: int | None = None,
+    loop_n: int | None = None,
+    all_duration: str | None = None,
    checkout: CheckoutOption = True,
    with_dashboard: bool = typer.Option(False, "--with-dashboard/-d", help="Start web dashboard automatically"),
    with_cli_dashboard: bool = typer.Option(False, "--cli-dashboard/-c", help="Show beautiful CLI dashboard"),
@@ -224,7 +230,7 @@ def fin_quant_cli(
        if not api_key:
            console.print("\n[bold red]❌ OPENROUTER_API_KEY not set in .env[/bold red]")
            console.print("[yellow]Add your API key to .env and retry:[/yellow]")
-            console.print('  OPENROUTER_API_KEY=sk-or-your-key-here')
+            console.print("  OPENROUTER_API_KEY=sk-or-your-key-here")
            raise typer.Exit(code=1)

        os.environ["OPENAI_API_KEY"] = api_key
@@ -243,8 +249,8 @@ def fin_quant_cli(
        console.print(f"   [dim]Base URL: {os.environ['OPENAI_API_BASE']}[/dim]")

        # Wait until the llama.cpp server is fully loaded before starting the pipeline
-        import urllib.request
        import urllib.error
+        import urllib.request

        base_url = os.environ["OPENAI_API_BASE"].removesuffix("/v1").rstrip("/")
        health_url = f"{base_url}/health"
@@ -278,7 +284,7 @@ def fin_quant_cli(
            subprocess.run(
                ["python", "web/dashboard_api.py"],
                cwd=str(Path(__file__).parent.parent.parent),
-                env={**os.environ, "FLASK_ENV": "development"}
+                env={**os.environ, "FLASK_ENV": "development"},
            )

        dashboard_thread = threading.Thread(target=start_web_dashboard, daemon=True)
@@ -320,9 +326,9 @@ def fin_quant_cli(

@app.command(name="fin_factor_report")
 def fin_factor_report_cli(
-    report_folder: Optional[str] = None,
-    path: Optional[str] = None,
-    all_duration: Optional[str] = None,
+    report_folder: str | None = None,
+    path: str | None = None,
+    all_duration: str | None = None,
    checkout: CheckoutOption = True,
 ):
    fin_factor_report(report_folder=report_folder, path=path, all_duration=all_duration, checkout=checkout)
@@ -335,12 +341,12 @@ def general_model_cli(report_file_path: str):

@app.command(name="data_science")
 def data_science_cli(
-    path: Optional[str] = None,
+    path: str | None = None,
    checkout: CheckoutOption = True,
-    step_n: Optional[int] = None,
-    loop_n: Optional[int] = None,
-    timeout: Optional[str] = None,
-    competition: Optional[str] = None,
+    step_n: int | None = None,
+    loop_n: int | None = None,
+    timeout: str | None = None,
+    competition: str | None = None,
 ):
    data_science(
        path=path,
@@ -354,16 +360,16 @@ def data_science_cli(

@app.command(name="llm_finetune")
 def llm_finetune_cli(
-    path: Optional[str] = None,
+    path: str | None = None,
    checkout: CheckoutOption = True,
-    benchmark: Optional[str] = None,
-    benchmark_description: Optional[str] = None,
-    dataset: Optional[str] = None,
-    base_model: Optional[str] = None,
-    upper_data_size_limit: Optional[int] = None,
-    step_n: Optional[int] = None,
-    loop_n: Optional[int] = None,
-    timeout: Optional[str] = None,
+    benchmark: str | None = None,
+    benchmark_description: str | None = None,
+    dataset: str | None = None,
+    base_model: str | None = None,
+    upper_data_size_limit: int | None = None,
+    step_n: int | None = None,
+    loop_n: int | None = None,
+    timeout: str | None = None,
 ):
    llm_finetune(
        path=path,
@@ -429,6 +435,7 @@ def rl_trading_cli(
        rdagent rl_trading --mode backtest --no-with-protections
    """
    from pathlib import Path
+
    import yaml

    console = Console()
@@ -440,18 +447,18 @@ def rl_trading_cli(
        with open(config_path) as f:
            config = yaml.safe_load(f) or {}

-    console.print(f"\n[bold blue]🤖 RL Trading Agent[/bold blue]")
+    console.print("\n[bold blue]🤖 RL Trading Agent[/bold blue]")
    console.print(f"Mode: [cyan]{mode}[/cyan]")
    console.print(f"Algorithm: [cyan]{algorithm.upper()}[/cyan]")
    console.print(f"Protections: {'[green]Enabled[/green]' if with_protections else '[red]Disabled[/red]'}")

    try:
-        from rdagent.components.coder.rl import RLTradingAgent, RLCosteer, TradingEnv
+        from rdagent.components.coder.rl import RLCosteer, RLTradingAgent, TradingEnv
    except ImportError as e:
-        console.print(f"[bold red]Error: RL components not available.[/bold red]")
+        console.print("[bold red]Error: RL components not available.[/bold red]")
        console.print(f"Details: {e}")
-        console.print(f"\n[yellow]Install RL dependencies:[/yellow]")
-        console.print(f"  pip install stable-baselines3 gymnasium")
+        console.print("\n[yellow]Install RL dependencies:[/yellow]")
+        console.print("  pip install stable-baselines3 gymnasium")
        raise typer.Exit(code=1)

    if mode == "train":
@@ -467,8 +474,8 @@ def rl_trading_cli(
            console.print("[dim]Loading market data...[/dim]")
            # TODO: Load actual data from config
            # For now, create mock environment
-            import numpy as np
            import gymnasium as gym
+            import numpy as np

            # Create simple mock environment for demonstration
            class MockTradingEnv(gym.Env):
@@ -504,7 +511,7 @@ def rl_trading_cli(
            model_path_out.parent.mkdir(parents=True, exist_ok=True)
            agent.save(model_path_out)

-            console.print(f"\n[bold green]✅ Training complete![/bold green]")
+            console.print("\n[bold green]✅ Training complete![/bold green]")
            console.print(f"Model saved to: [cyan]{model_path_out}[/cyan]")
            console.print(f"Algorithm: {result['algorithm']}")
            console.print(f"Timesteps: {result['total_timesteps']:,}")
@@ -530,9 +537,9 @@ def rl_trading_cli(
                agent = RLTradingAgent(algorithm=algorithm.upper())

            # Run backtest
-            from rdagent.components.backtesting import FactorBacktester
-            import pandas as pd
            import numpy as np
+            import pandas as pd
+            from rdagent.components.backtesting import FactorBacktester

            backtester = FactorBacktester()

@@ -541,8 +548,8 @@ def rl_trading_cli(
            n_steps = 500
            mock_prices = pd.Series(100 + np.cumsum(np.random.randn(n_steps) * 0.5))
            mock_indicators = pd.DataFrame({
-                'rsi': np.random.uniform(30, 70, n_steps),
-                'macd': np.random.randn(n_steps) * 0.1,
+                "rsi": np.random.uniform(30, 70, n_steps),
+                "macd": np.random.randn(n_steps) * 0.1,
            })

            console.print("[yellow]Running backtest...[/yellow]")
@@ -553,7 +560,7 @@ def rl_trading_cli(
                enable_protections=with_protections,
            )

-            console.print(f"\n[bold green]✅ Backtest complete![/bold green]")
+            console.print("\n[bold green]✅ Backtest complete![/bold green]")
            console.print(f"  Final Equity: [green]${metrics.get('final_equity', 0):,.2f}[/green]")
            console.print(f"  Sharpe Ratio: {metrics.get('sharpe_ratio', 0):.3f}")
            console.print(f"  Max Drawdown: {metrics.get('max_drawdown', 0):.2%}")
@@ -627,7 +634,7 @@ def generate_strategies_cli(
        rdagent generate_strategies -n 3 -i 10 --optuna-trials 50  # Deep optimization
    """
    from rich.console import Console
-    from rich.progress import Progress, SpinnerColumn, TextColumn, BarColumn, TimeRemainingColumn
+    from rich.progress import BarColumn, Progress, SpinnerColumn, TextColumn, TimeRemainingColumn
    from rich.table import Table

    console = Console()
@@ -646,7 +653,7 @@ def generate_strategies_cli(
        raise typer.Exit(code=1)

    console.print(f"\n[bold blue]{'='*60}[/bold blue]")
-    console.print(f"[bold blue]  PREDIX Strategy Generator[/bold blue]")
+    console.print("[bold blue]  PREDIX Strategy Generator[/bold blue]")
    console.print(f"[bold blue]{'='*60}[/bold blue]")
    console.print(f"  Strategies:  [cyan]{count}[/cyan]")
    console.print(f"  Workers:     [cyan]{workers}[/cyan]")
@@ -673,12 +680,12 @@ def generate_strategies_cli(
    _slog = _dlog.setup("strategies", **_strat_ctx)

    try:
-        from rdagent.components.coder.strategy_orchestrator import StrategyOrchestrator
        import pandas as pd
+        from rdagent.components.coder.strategy_orchestrator import StrategyOrchestrator

        all_results = []
        best_strategy = None
-        best_sharpe = float('-inf')
+        best_sharpe = float("-inf")

        # CONTINUOUS OPTIMIZATION LOOP
        for iteration in range(1, max_iterations + 1):
@@ -727,7 +734,7 @@ def generate_strategies_cli(

            # Track best strategy
            for r in results:
-                sharpe = r.get("sharpe_ratio", float('-inf'))
+                sharpe = r.get("sharpe_ratio", float("-inf"))
                if sharpe > best_sharpe:
                    best_sharpe = sharpe
                    best_strategy = r
@@ -747,7 +754,7 @@ def generate_strategies_cli(
        rejected = [r for r in results if r.get("status") == "rejected"]

        console.print(f"\n[bold green]{'='*60}[/bold green]")
-        console.print(f"[bold green]  Strategy Generation Summary[/bold green]")
+        console.print("[bold green]  Strategy Generation Summary[/bold green]")
        console.print(f"[bold green]{'='*60}[/bold green]")

        table = Table(show_header=True, header_style="bold magenta", show_lines=True)
@@ -776,7 +783,7 @@ def generate_strategies_cli(
        # Show best strategy details
        if best_strategy:
            console.print(f"\n[bold gold1]{'='*60}[/bold gold1]")
-            console.print(f"[bold gold1]  BEST STRATEGY[/bold gold1]")
+            console.print("[bold gold1]  BEST STRATEGY[/bold gold1]")
            console.print(f"[bold gold1]{'='*60}[/bold gold1]")
            console.print(f"  Name:        [cyan]{best_strategy.get('strategy_name', 'Unknown')}[/cyan]")
            console.print(f"  Sharpe:      [green]{best_strategy.get('sharpe_ratio', 0):.4f}[/green]")
@@ -784,13 +791,13 @@ def generate_strategies_cli(
            console.print(f"  Max DD:      [yellow]{best_strategy.get('max_drawdown', 0):.2%}[/yellow]")
            console.print(f"  Win Rate:    [cyan]{best_strategy.get('win_rate', 0):.2%}[/cyan]")
            if best_strategy.get("best_params"):
-                console.print(f"\n  [bold]Optimized Parameters:[/bold]")
+                console.print("\n  [bold]Optimized Parameters:[/bold]")
                for param, val in best_strategy["best_params"].items():
                    console.print(f"    {param}: [cyan]{val}[/cyan]")
            console.print(f"[bold gold1]{'='*60}[/bold gold1]")

        if accepted:
-            console.print(f"\n[bold]Accepted Strategies:[/bold]")
+            console.print("\n[bold]Accepted Strategies:[/bold]")
            acc_table = Table(show_header=True, header_style="bold cyan")
            acc_table.add_column("#", width=4)
            acc_table.add_column("Strategy", width=30)
@@ -813,13 +820,13 @@ def generate_strategies_cli(
                )
            console.print(acc_table)

-        console.print(f"\n[bold green]Strategies saved to:[/bold green] [cyan]results/strategies_new/[/cyan]")
+        console.print("\n[bold green]Strategies saved to:[/bold green] [cyan]results/strategies_new/[/cyan]")
        console.print(f"[bold blue]{'='*60}[/bold blue]\n")
        _slog.success(f"Generated {len(all_results)} strategies ({len([r for r in all_results if r.get('status')=='accepted'])} accepted)")

    except ImportError as e:
        _slog.error(f"Strategy components not available: {e}")
-        console.print(f"[bold red]Error: Strategy components not available.[/bold red]")
+        console.print("[bold red]Error: Strategy components not available.[/bold red]")
        console.print(f"Details: {e}")
        raise typer.Exit(code=1)
    except Exception as e:
@@ -855,17 +862,18 @@ def optimize_portfolio_cli(
        raise typer.Exit(code=1)

    console.print(f"\n[bold blue]{'='*60}[/bold blue]")
-    console.print(f"[bold blue]  PREDIX Portfolio Optimizer[/bold blue]")
+    console.print("[bold blue]  PREDIX Portfolio Optimizer[/bold blue]")
    console.print(f"[bold blue]{'='*60}[/bold blue]")
    console.print(f"  Top N:    [cyan]{top_n}[/cyan]")
    console.print(f"  Method:   [cyan]{method}[/cyan]")
    console.print(f"[bold blue]{'='*60}[/bold blue]\n")

    try:
-        from rdagent.components.backtesting.risk_management import PortfolioOptimizer
        import json
        from pathlib import Path

+        from rdagent.components.backtesting.risk_management import PortfolioOptimizer
+
        project_root = Path(__file__).parent.parent.parent
        strategies_dir = project_root / "results" / "strategies_new"

@@ -882,6 +890,7 @@ def optimize_portfolio_cli(
                if data.get("status") == "accepted":
                    strategies.append(data)
            except Exception:
+                logger.warning("Failed to load strategy file %s", f, exc_info=True)
                continue

        if not strategies:
@@ -1001,14 +1010,15 @@ def strategies_report_cli(
        rdagent strategies_report -s path/to/strategy.json  # Single strategy
        rdagent strategies_report -o custom/reports/      # Custom output dir
    """
+    from pathlib import Path
+
    from rich.console import Console
    from rich.progress import Progress, SpinnerColumn, TextColumn
-    from pathlib import Path

    console = Console()

    console.print(f"\n[bold blue]{'='*60}[/bold blue]")
-    console.print(f"[bold blue]  PREDIX Strategy Report Generator[/bold blue]")
+    console.print("[bold blue]  PREDIX Strategy Report Generator[/bold blue]")
    console.print(f"[bold blue]{'='*60}[/bold blue]\n")

    project_root = Path(__file__).parent.parent.parent
@@ -1060,20 +1070,20 @@ def strategies_report_cli(
                progress.update(task, completed=1)

    console.print(f"\n[bold green]{'='*60}[/bold green]")
-    console.print(f"[bold green]  Report Generation Complete[/bold green]")
+    console.print("[bold green]  Report Generation Complete[/bold green]")
    console.print(f"[bold green]{'='*60}[/bold green]")
    console.print(f"  Reports generated: [cyan]{reports_generated}/{len(strategy_files)}[/cyan]")
    console.print(f"  Output directory:  [cyan]{output_dir_path}[/cyan]")
    console.print(f"[bold green]{'='*60}[/bold green]\n")


-def _generate_single_strategy_report(strategy_file: Path, output_dir: Path) -> Dict:
+def _generate_single_strategy_report(strategy_file: Path, output_dir: Path) -> dict:
    """Generate a report for a single strategy."""
    import json
+
    import matplotlib
    matplotlib.use("Agg")  # Non-interactive backend
    import matplotlib.pyplot as plt
-    import seaborn as sns

    with open(strategy_file, encoding="utf-8") as f:
        strategy = json.load(f)
@@ -1148,7 +1158,7 @@ if __name__ == "__main__":
@app.command(name="start_llama")
 def start_llama_cli(
    model: str = typer.Option(
-        None, "--model", "-m", help="Path to model file"
+        None, "--model", "-m", help="Path to model file",
    ),
    port: int = typer.Option(8081, "--port", "-p", help="Server port"),
    gpu_layers: int = typer.Option(30, "--gpu-layers", "-g", help="GPU layers"),
@@ -1170,8 +1180,6 @@ def start_llama_cli(
      rdagent start_llama --gpu-layers 40 --ctx-size 4096
      rdagent start_llama --reasoning
    """
-    import subprocess
-    import sys
    import os

    model_path = model or os.getenv(
@@ -1208,7 +1216,7 @@ def start_llama_cli(
    if not reasoning:
        cmd.extend(["--reasoning", "off"])

-    print(f"🚀 Starting llama.cpp server...")
+    print("🚀 Starting llama.cpp server...")
    print(f"   Model: {Path(model_path).name}")
    print(f"   Port: {port}")
    print(f"   GPU Layers: {gpu_layers}")
@@ -1241,17 +1249,16 @@ def start_loop_cli(
      rdagent start_loop
      rdagent start_loop --target 5 --max-wait 3600
    """
-    import subprocess
-    import signal
-    import sys
    import os
-    from datetime import datetime
+    import signal
+    import subprocess
    import time
+    from datetime import datetime

-    script_dir = str(Path(__file__).parent.parent.parent.parent)
-    generator = f"python {script_dir}/scripts/predix_smart_strategy_gen.py"
+    script_dir = str(Path(__file__).parent.parent.parent)
+    generator = [sys.executable, f"{script_dir}/scripts/predix_smart_strategy_gen.py"]
    logfile = f"{script_dir}/results/logs/generator_loop.log"
-    pidfile = "/tmp/predix_loop.pid"
+    pidfile = "/tmp/predix_loop.pid"  # nosec B108 — administrative PID file, single-process daemon

    os.makedirs(f"{script_dir}/results/logs", exist_ok=True)

@@ -1262,12 +1269,19 @@ def start_loop_cli(
        with open(logfile, "a") as f:
            f.write(line + "\n")

+    child_proc = None  # track current child PID for targeted cleanup
+
    def cleanup(signum=None, frame=None):
        log("Received termination signal. Cleaning up...")
-        try:
-            subprocess.run(["pkill", "-f", "predix_smart_strategy_gen.py"], capture_output=True)
-        except Exception:
-            pass
+        if child_proc is not None:
+            try:
+                child_proc.terminate()
+                child_proc.wait(timeout=10)
+            except Exception:
+                try:
+                    child_proc.kill()
+                except Exception:
+                    pass
        try:
            os.remove(pidfile)
        except FileNotFoundError:
@@ -1313,26 +1327,32 @@ def start_loop_cli(
        strat_count = len(list(strat_dir.glob("*.json"))) if strat_dir.exists() else 0
        log(f"📁 Existing strategies: {strat_count}")

-        # Kill stale processes
-        try:
-            subprocess.run(["pkill", "-9", "-f", "predix_smart_strategy_gen.py"], capture_output=True)
-        except Exception:
-            pass
-        time.sleep(2)
+        # Kill stale child from previous iteration
+        if child_proc is not None:
+            try:
+                child_proc.terminate()
+                child_proc.wait(timeout=10)
+            except subprocess.TimeoutExpired:
+                child_proc.kill()
+                child_proc.wait()
+            except Exception:
+                pass
+            child_proc = None
+            time.sleep(2)

        # Start generator
        log("🤖 Starting generator...")
-        proc = subprocess.Popen(
-            generator.split(),
+        child_proc = subprocess.Popen(
+            generator,
            cwd=script_dir,
            stdout=subprocess.DEVNULL,
            stderr=subprocess.DEVNULL,
        )
-        log(f"   PID: {proc.pid}")
+        log(f"   PID: {child_proc.pid}")

        # Monitor progress
        elapsed = 0
-        while proc.poll() is None:
+        while child_proc.poll() is None:
            time.sleep(30)
            elapsed += 30

@@ -1341,11 +1361,12 @@ def start_loop_cli(

            if elapsed >= max_wait:
                log(f"   ⏰ Timeout after {elapsed}s. Killing...")
-                proc.kill()
+                child_proc.kill()
                break

        # Check results
-        exit_code = proc.wait()
+        exit_code = child_proc.wait()
+        child_proc = None
        if exit_code == 0:
            log("✅ Generator completed successfully")
        elif exit_code == -9:
@@ -1386,24 +1407,24 @@ def parallel_cli(
      rdagent parallel -n 10 -k 2
    """
    import subprocess
-    import sys
    from pathlib import Path
+
    from rdagent.log import daily_log as _dlog

-    project_root = Path(__file__).parent.parent.parent.parent
+    project_root = Path(__file__).parent.parent.parent
    script = project_root / "scripts" / "predix_parallel.py"

    if not script.exists():
        typer.echo(f"❌ Script not found: {script}")
        raise typer.Exit(code=1)

-    cmd = [sys.executable, str(script), "--runs", str(runs), "--api-keys", str(api_keys), "-m", "local"]
+    cmd = [sys.executable, str(script), "--runs", str(runs), "--api-keys", str(api_keys)]

    _plog = _dlog.setup("parallel", runs=runs, api_keys=api_keys, model="local")
    typer.echo(f"🚀 Starting {runs} parallel runs...")
    typer.echo(f"   Script: {script}")
    typer.echo(f"   API Keys: {api_keys}")
-    typer.echo(f"   Model: local (llama.cpp)")
+    typer.echo("   Model: local (llama.cpp)")

    try:
        result = subprocess.run(cmd, cwd=str(project_root))
@@ -1437,11 +1458,11 @@ def eval_all_cli(
      rdagent eval_all -n 500 -p 8
    """
    import subprocess
-    import sys
    from pathlib import Path
+
    from rdagent.log import daily_log as _dlog

-    project_root = Path(__file__).parent.parent.parent.parent
+    project_root = Path(__file__).parent.parent.parent
    script = project_root / "scripts" / "predix_full_eval.py"

    if not script.exists():
@@ -1492,10 +1513,9 @@ def batch_backtest_cli(
      rdagent batch_backtest --all
    """
    import subprocess
-    import sys
    from pathlib import Path

-    project_root = Path(__file__).parent.parent.parent.parent
+    project_root = Path(__file__).parent.parent.parent
    script = project_root / "scripts" / "predix_batch_backtest.py"

    if not script.exists():
@@ -1545,10 +1565,9 @@ def simple_eval_cli(
      rdagent simple_eval --all
    """
    import subprocess
-    import sys
    from pathlib import Path

-    project_root = Path(__file__).parent.parent.parent.parent
+    project_root = Path(__file__).parent.parent.parent
    script = project_root / "scripts" / "predix_simple_eval.py"

    if not script.exists():
@@ -1578,7 +1597,7 @@ def simple_eval_cli(
@app.command(name="rebacktest")
 def rebacktest_cli(
    strategies_dir: str = typer.Option(
-        None, "--strategies-dir", "-d", help="Directory containing strategy JSON files"
+        None, "--strategies-dir", "-d", help="Directory containing strategy JSON files",
    ),
 ):
    """
@@ -1592,10 +1611,9 @@ def rebacktest_cli(
      rdagent rebacktest -d results/strategies_new/
    """
    import subprocess
-    import sys
    from pathlib import Path

-    project_root = Path(__file__).parent.parent.parent.parent
+    project_root = Path(__file__).parent.parent.parent
    script = project_root / "scripts" / "predix_rebacktest_strategies.py"

    if not script.exists():
@@ -1620,10 +1638,10 @@ def rebacktest_cli(
@app.command(name="report")
 def report_cli(
    strategy_path: str = typer.Option(
-        None, "--strategy", "-s", help="Path to single strategy JSON (default: all strategies)"
+        None, "--strategy", "-s", help="Path to single strategy JSON (default: all strategies)",
    ),
    output: str = typer.Option(
-        None, "--output", "-o", help="Output directory (default: results/strategy_reports/)"
+        None, "--output", "-o", help="Output directory (default: results/strategy_reports/)",
    ),
 ):
    """
@@ -1646,10 +1664,9 @@ def report_cli(
      rdagent report -o custom/reports/
    """
    import subprocess
-    import sys
    from pathlib import Path

-    project_root = Path(__file__).parent.parent.parent.parent
+    project_root = Path(__file__).parent.parent.parent
    script = project_root / "scripts" / "predix_strategy_report.py"

    if not script.exists():
@@ -201,6 +201,5 @@ class DataScienceBasePropSetting(KaggleBasePropSetting):
 DS_RD_SETTING = DataScienceBasePropSetting()

 # enable_cross_trace_diversity and llm_select_hypothesis should not be true at the same time
-assert not (
-    DS_RD_SETTING.enable_cross_trace_diversity and DS_RD_SETTING.llm_select_hypothesis
-), "enable_cross_trace_diversity and llm_select_hypothesis cannot be true at the same time"
+if DS_RD_SETTING.enable_cross_trace_diversity and DS_RD_SETTING.llm_select_hypothesis:
+    raise ValueError("enable_cross_trace_diversity and llm_select_hypothesis cannot be true at the same time")
@@ -58,18 +58,18 @@ def main(

    if user_target_scenario:
        FT_RD_SETTING.user_target_scenario = user_target_scenario
-    assert (
-        FT_RD_SETTING.user_target_scenario is None
-    ), "user_target_scenario is not yet supported, please specify via benchmark and benchmark_description"
+    if FT_RD_SETTING.user_target_scenario is not None:
+        raise ValueError("user_target_scenario is not yet supported, please specify via benchmark and benchmark_description")
    if upper_data_size_limit:
        FT_RD_SETTING.upper_data_size_limit = upper_data_size_limit
        logger.info(f"Set upper_data_size_limit to {FT_RD_SETTING.upper_data_size_limit}")
    if benchmark and benchmark_description:
        FT_RD_SETTING.target_benchmark = benchmark
        FT_RD_SETTING.benchmark_description = benchmark_description
-    assert FT_RD_SETTING.user_target_scenario or (
-        FT_RD_SETTING.target_benchmark and FT_RD_SETTING.benchmark_description
-    ), "Either user_target_scenario or target_benchmark must be specified for LLM fine-tuning."
+    if not (
+        FT_RD_SETTING.user_target_scenario or (FT_RD_SETTING.target_benchmark and FT_RD_SETTING.benchmark_description)
+    ):
+        raise ValueError("Either user_target_scenario or target_benchmark must be specified for LLM fine-tuning.")

    # Update configuration with provided parameters
    if dataset:
@@ -82,9 +82,8 @@ def main(
    model_target = FT_RD_SETTING.base_model if FT_RD_SETTING.base_model else "auto selected model"

    # Temporary assertion until auto-selection is implemented
-    assert (
-        FT_RD_SETTING.base_model is not None
-    ), "Base model auto selection not yet supported, please specify via --base-model"
+    if FT_RD_SETTING.base_model is None:
+        raise ValueError("Base model auto selection not yet supported, please specify via --base-model")

    logger.info(f"Starting LLM fine-tuning on dataset='{data_set_target}' with model='{model_target}'")

@@ -24,46 +24,12 @@ from rdagent.app.finetune.llm.ui.ft_summary import render_job_summary

 DEFAULT_LOG_BASE = "log/"

+from rdagent.core.utils import safe_resolve_path
+

 def validate_path_within_cwd(user_path: Path) -> Path:
-    """
-    Validate that a user-provided path is within the current working directory.
-
-    Security: This function prevents path traversal attacks by:
-    1. Resolving the path to its absolute canonical form
-    2. Verifying it's within the CWD boundary using a normalized common prefix
-    3. Rejecting paths outside the boundary with ValueError
-
-    Parameters
-    ----------
-    user_path : Path
-        User-provided path to validate
-
-    Returns
-    -------
-    Path
-        Resolved absolute path if valid
-
-    Raises
-    ------
-    ValueError
-        If path is outside the current working directory
-    """
    safe_root = Path.cwd().resolve()
-    # Expand any user home reference and resolve without requiring the path to exist.
-    resolved_path = user_path.expanduser().resolve(strict=False)
-
-    # Ensure the resolved path is absolute and remains within the safe root.
-    safe_root_str = str(safe_root)
-    resolved_str = str(resolved_path)
-    common = os.path.commonpath([safe_root_str, resolved_str])
-    if common != safe_root_str:
-        raise ValueError("Path is outside the allowed project directory")
-
-    # This will raise ValueError if resolved_path is not within safe_root
-    resolved_path.relative_to(safe_root)
-
-    return resolved_path
+    return safe_resolve_path(user_path, safe_root)


 def get_job_options(base_path: Path, safe_root: Path | None = None) -> list[str]:
@@ -141,19 +107,14 @@ def main():
            st.header("Job")
            base_folder = st.text_input("Base Folder", value=default_log, key="base_folder_input")

-            # Normalize and validate the base folder against the configured log root
-            root_real = os.path.realpath(str(Path(default_log).expanduser()))
-            folder_real = os.path.realpath(str(Path(base_folder).expanduser()))
-            if folder_real == root_real or folder_real.startswith(root_real + os.sep):
-                base_path = Path(folder_real)
-                safe_root = Path(root_real)
-            else:
+            safe_root = Path(default_log).expanduser().resolve()
+            try:
+                base_path = safe_resolve_path(Path(base_folder), safe_root)
+            except ValueError:
                st.error("Invalid base folder: must be within the configured log directory.")
-                safe_root = Path(root_real)
                base_path = safe_root

-            # base_path is validated against safe_root – nosec B614
-            job_options = get_job_options(base_path, safe_root)  # nosec B614 – validated above
+            job_options = get_job_options(base_path, safe_root)
            if job_options:
                selected_job = st.selectbox("Select Job", job_options, key="job_select")
                if selected_job.startswith("."):
@@ -13,6 +13,7 @@ from typing import Any
 import streamlit as st

 from rdagent.app.finetune.llm.ui.config import EVALUATOR_CONFIG, EventType
+from rdagent.core.utils import safe_resolve_path
 from rdagent.log.storage import FileStorage


@@ -89,11 +90,10 @@ def extract_stage(tag: str) -> str:
 def get_valid_sessions(log_folder: Path, safe_root: Path | None = None) -> list[str]:
    """Get list of valid session directories, optionally validating against a safe root."""
    if safe_root is not None:
-        root_real = os.path.realpath(str(safe_root.expanduser()))
-        folder_real = os.path.realpath(str(log_folder.expanduser()))
-        if not (folder_real == root_real or folder_real.startswith(root_real + os.sep)):
+        try:
+            log_folder = safe_resolve_path(log_folder, safe_root)
+        except ValueError:
            return []
-        log_folder = Path(folder_real)

    if not log_folder.exists():
        return []
@@ -373,13 +373,11 @@ def parse_event(tag: str, content: Any, timestamp: datetime) -> Event | None:
@st.cache_data(ttl=300, hash_funcs={Path: str})
 def load_ft_session(log_path: Path, safe_root: Path | None = None) -> Session:
    """Load events into hierarchical session structure, optionally validating against safe root."""
-    # Validate path is within safe_root if provided
    if safe_root is not None:
-        root_real = os.path.realpath(str(safe_root.expanduser()))
-        path_real = os.path.realpath(str(log_path.expanduser()))
-        if not (path_real == root_real or path_real.startswith(root_real + os.sep)):
+        try:
+            log_path = safe_resolve_path(log_path, safe_root)
+        except ValueError:
            return Session()
-        log_path = Path(path_real)

    session = Session()
    storage = FileStorage(log_path)
@@ -4,10 +4,9 @@ Factor workflow with session control

 import asyncio
 from pathlib import Path
-from typing import Any, Optional
+from typing import Any

 import fire
-
 from rdagent.app.qlib_rd_loop.conf import FACTOR_PROP_SETTING
 from rdagent.components.workflow.rd_loop import RDLoop
 from rdagent.core.exception import CoderError, FactorEmptyError
@@ -21,20 +20,20 @@ class FactorRDLoop(RDLoop):
    def running(self, prev_out: dict[str, Any]):
        exp = self.runner.develop(prev_out["coding"])
        if exp is None:
-            logger.error(f"Factor extraction failed.")
+            logger.error("Factor extraction failed.")
            raise FactorEmptyError("Factor extraction failed.")
        logger.log_object(exp, tag="runner result")
        return exp


 def main(
-    path: Optional[str] = None,
-    step_n: Optional[int] = None,
-    loop_n: Optional[int] = None,
+    path: str | None = None,
+    step_n: int | None = None,
+    loop_n: int | None = None,
    all_duration: str | None = None,
    checkout: bool = True,
-    checkout_path: Optional[str] = None,
-    base_features_path: Optional[str] = None,
+    checkout_path: str | None = None,
+    base_features_path: str | None = None,
    **kwargs,
 ):
    """
@@ -47,7 +46,7 @@ def main(
        dotenv run -- python rdagent/app/qlib_rd_loop/factor.py $LOG_PATH/__session__/1/0_propose  --step_n 1   # `step_n` is a optional paramter

    """
-    if not checkout_path is None:
+    if checkout_path is not None:
        checkout = Path(checkout_path)

    if path is None:
@@ -1,10 +1,9 @@
 import asyncio
 import json
 from pathlib import Path
-from typing import Any, Dict, Tuple
+from typing import Any

 import fire
-
 from rdagent.app.qlib_rd_loop.conf import FACTOR_FROM_REPORT_PROP_SETTING
 from rdagent.app.qlib_rd_loop.factor import FactorRDLoop
 from rdagent.components.document_reader.document_reader import (
@@ -12,7 +11,7 @@ from rdagent.components.document_reader.document_reader import (
    load_and_process_pdfs_by_langchain,
 )
 from rdagent.core.conf import RD_AGENT_SETTINGS
-from rdagent.core.proposal import Hypothesis, HypothesisFeedback
+from rdagent.core.proposal import Hypothesis
 from rdagent.log import rdagent_logger as logger
 from rdagent.oai.llm_utils import APIBackend
 from rdagent.scenarios.qlib.experiment.factor_experiment import QlibFactorExperiment
@@ -36,14 +35,14 @@ def generate_hypothesis(factor_result: dict, report_content: str) -> str:
    """
    system_prompt = T(".prompts:hypothesis_generation.system").r()
    user_prompt = T(".prompts:hypothesis_generation.user").r(
-        factor_descriptions=json.dumps(factor_result), report_content=report_content
+        factor_descriptions=json.dumps(factor_result), report_content=report_content,
    )

    response = APIBackend().build_messages_and_create_chat_completion(
        user_prompt=user_prompt,
        system_prompt=system_prompt,
        json_mode=True,
-        json_target_type=Dict[str, str],
+        json_target_type=dict[str, str],
    )

    response_json = json.loads(response)
@@ -99,7 +98,7 @@ class FactorReportLoop(FactorRDLoop, metaclass=LoopMeta):
        super().__init__(PROP_SETTING=FACTOR_FROM_REPORT_PROP_SETTING)
        if report_folder is None:
            self.judge_pdf_data_items = json.load(
-                open(FACTOR_FROM_REPORT_PROP_SETTING.report_result_json_file_path, "r")
+                open(FACTOR_FROM_REPORT_PROP_SETTING.report_result_json_file_path),
            )
        else:
            self.judge_pdf_data_items = [i for i in Path(report_folder).rglob("*.pdf")]
@@ -118,7 +117,7 @@ class FactorReportLoop(FactorRDLoop, metaclass=LoopMeta):
                if exp is None:
                    self.shift_report += 1
                    self.loop_n -= 1
-                    if self.loop_n < 0:  # NOTE: on every step, we self.loop_n -= 1 at first.
+                    if self.loop_n < 0:  # loop_n is decremented above when reports are empty; prevents infinite skipping
                        raise self.LoopTerminationError("Reach stop criterion and stop loop")
                    continue
                exp.based_experiments = [QlibFactorExperiment(sub_tasks=[], hypothesis=exp.hypothesis)] + [
@@ -8,7 +8,6 @@ from pathlib import Path
 from typing import Any

 import fire
-
 from rdagent.app.qlib_rd_loop.conf import QUANT_PROP_SETTING
 from rdagent.components.workflow.conf import BasePropSetting
 from rdagent.components.workflow.rd_loop import RDLoop
@@ -44,11 +43,11 @@ class QuantRDLoop(RDLoop):
        logger.log_object(self.hypothesis_gen, tag="quant hypothesis generator")

        self.factor_hypothesis2experiment: Hypothesis2Experiment = import_class(
-            PROP_SETTING.factor_hypothesis2experiment
+            PROP_SETTING.factor_hypothesis2experiment,
        )()
        logger.log_object(self.factor_hypothesis2experiment, tag="factor hypothesis2experiment")
        self.model_hypothesis2experiment: Hypothesis2Experiment = import_class(
-            PROP_SETTING.model_hypothesis2experiment
+            PROP_SETTING.model_hypothesis2experiment,
        )()
        logger.log_object(self.model_hypothesis2experiment, tag="model hypothesis2experiment")

@@ -78,7 +77,8 @@ class QuantRDLoop(RDLoop):
        while True:
            if self.get_unfinished_loop_cnt(self.loop_idx) < RD_AGENT_SETTINGS.get_max_parallel():
                hypo = self._propose()
-                assert hypo.action in ["factor", "model"]
+                if hypo.action not in ["factor", "model"]:
+                    raise ValueError(f"hypo.action must be 'factor' or 'model', got {hypo.action!r}")
                if hypo.action == "factor":
                    exp = self.factor_hypothesis2experiment.convert(hypo, self.trace)
                else:
@@ -132,7 +132,6 @@ class QuantRDLoop(RDLoop):
        """
        import json
        from datetime import datetime
-        from pathlib import Path

        try:
            project_root = Path(__file__).parent.parent.parent.parent
@@ -195,11 +194,11 @@ class QuantRDLoop(RDLoop):
        if prev_out["direct_exp_gen"]["propose"].action == "factor":
            exp = self.factor_runner.develop(prev_out["coding"])
            if exp is None:
-                logger.error(f"Factor extraction failed.")
+                logger.error("Factor extraction failed.")
                raise FactorEmptyError("Factor extraction failed.")

            # Increment factor count for tracking
-            if hasattr(self, 'trace') and hasattr(self.trace, 'increment_factor_count'):
+            if hasattr(self, "trace") and hasattr(self.trace, "increment_factor_count"):
                self.trace.increment_factor_count()

            # Handle failed experiments gracefully (don't break the loop)
@@ -210,7 +209,7 @@ class QuantRDLoop(RDLoop):
                    factor_name = getattr(exp.hypothesis, "hypothesis", "unknown")
                logger.warning(
                    f"Factor '{factor_name}' failed evaluation: {reason}. "
-                    f"Continuing with next factor."
+                    f"Continuing with next factor.",
                )
                # Return exp anyway - loop will continue
        elif prev_out["direct_exp_gen"]["propose"].action == "model":
@@ -219,7 +218,7 @@ class QuantRDLoop(RDLoop):
        return exp

    def feedback(self, prev_out: dict[str, Any]):
-        e = prev_out.get(self.EXCEPTION_KEY, None)
+        e = prev_out.get(self.EXCEPTION_KEY)
        if e is not None:
            feedback = HypothesisFeedback(
                observations=str(e),
@@ -245,11 +244,10 @@ class QuantRDLoop(RDLoop):
                    reason=reason,
                    decision=False,
                )
-            else:
-                if prev_out["direct_exp_gen"]["propose"].action == "factor":
-                    feedback = self.factor_summarizer.generate_feedback(prev_out["running"], self.trace)
-                elif prev_out["direct_exp_gen"]["propose"].action == "model":
-                    feedback = self.model_summarizer.generate_feedback(prev_out["running"], self.trace)
+            elif prev_out["direct_exp_gen"]["propose"].action == "factor":
+                feedback = self.factor_summarizer.generate_feedback(prev_out["running"], self.trace)
+            elif prev_out["direct_exp_gen"]["propose"].action == "model":
+                feedback = self.model_summarizer.generate_feedback(prev_out["running"], self.trace)

            # NOTE: DB save is handled by factor_runner.py _save_result_to_database()
            # which runs immediately after Docker execution. No duplicate save needed here.
@@ -258,20 +256,20 @@ class QuantRDLoop(RDLoop):
        factor_count = self.trace.get_factor_count()

        # Check for auto-strategies trigger
-        auto_strategies = getattr(self, '_auto_strategies', False)
-        auto_threshold = getattr(self, '_auto_strategies_threshold', 500)
+        auto_strategies = getattr(self, "_auto_strategies", False)
+        auto_threshold = getattr(self, "_auto_strategies_threshold", 500)

        if auto_strategies and factor_count > 0 and factor_count % auto_threshold == 0:
            logger.info(
                f"Auto-strategy trigger: {factor_count} factors evaluated. "
-                f"Suggesting strategy generation now..."
+                f"Suggesting strategy generation now...",
            )
            self._build_strategies_with_ai()
        elif factor_count > 0 and factor_count % 50 == 0 and not auto_strategies:
            # Standard periodic suggestion (every 50 factors)
            logger.info(
                f"Periodic check: {factor_count} factors evaluated. "
-                f"Consider running 'rdagent generate_strategies' for AI strategy generation."
+                f"Consider running 'rdagent generate_strategies' for AI strategy generation.",
            )

        feedback = self._interact_feedback(feedback)
@@ -292,10 +290,11 @@ class QuantRDLoop(RDLoop):
        - Optuna hyperparameter optimization
        """
        try:
-            from rdagent.components.coder.strategy_orchestrator import StrategyOrchestrator
            from pathlib import Path
+
            import yaml
-            
+            from rdagent.components.coder.strategy_orchestrator import StrategyOrchestrator
+
            # Load improved prompt
            project_root = Path(__file__).parent.parent.parent.parent
            prompt_path = project_root / "prompts" / "strategy_generation_v2.yaml"
@@ -322,6 +321,7 @@ class QuantRDLoop(RDLoop):
                    if data.get("status") == "success" and data.get("ic") is not None:
                        factors.append(data)
                except Exception:
+                    logger.warning("Failed to load factor file %s", f, exc_info=True)
                    continue

            if len(factors) < 10:
@@ -334,44 +334,47 @@ class QuantRDLoop(RDLoop):

            logger.info(f"StrategyOrchestrator: Building strategies from {len(top_factors)} top factors...")
            logger.info(f"  - Using improved prompt: {improved_prompt is not None}")
-            logger.info(f"  - Optuna optimization: enabled (20 trials)")
-            logger.info(f"  - Real OHLCV backtest: enabled")
+            logger.info("  - Optuna optimization: enabled (20 trials)")
+            logger.info("  - Real OHLCV backtest: enabled")

            # Initialize orchestrator with Optuna
            orchestrator = StrategyOrchestrator(
                top_factors=20,
-                trading_style='swing',
+                trading_style="swing",
                min_sharpe=0.5,
                max_drawdown=-0.20,
                min_win_rate=0.40,
                use_optuna=True,
                optuna_trials=20,
            )
-            
+
            # Override with improved prompt if available
            if improved_prompt:
-                orchestrator.strategy_prompt = improved_prompt.get('strategy_generation', {})
+                orchestrator.strategy_prompt = improved_prompt.get("strategy_generation", {})

            # Generate 3 strategies per cycle
            n_strategies = 3
            logger.info(f"Generating {n_strategies} strategies...")
-            
+
            # Load top factors for generation
            orch_factors = orchestrator.load_top_factors()
-            
+            if len(orch_factors) < 2:
+                logger.warning(f"Not enough factors for strategy generation (need >= 2, got {len(orch_factors)}). Skipping.")
+                return
+
            for i in range(n_strategies):
+                strategy_name = f"auto_gen_v{i+1}"
                try:
                    # Select random factor combination
                    import random
                    n_factors = random.randint(2, min(5, len(orch_factors)))
                    factor_subset = random.sample(orch_factors, n_factors)
-                    
-                    strategy_name = f"auto_gen_v{i+1}"
+
                    code = orchestrator.generate_strategy_code(factor_subset, strategy_name)
-                    
+
                    if code:
                        result = orchestrator.evaluate_strategy(code, strategy_name, factor_subset)
-                        
+
                        if result.get("status") == "accepted":
                            logger.info(f"✅ Strategy {strategy_name} accepted!")
                            logger.info(f"   Sharpe: {result.get('sharpe_ratio', 0):.2f}")
@@ -429,7 +432,7 @@ def main(
        quant_loop._auto_strategies = True
        quant_loop._auto_strategies_threshold = auto_strategies_threshold
        logger.info(
-            f"Auto-strategies enabled. Will trigger after {auto_strategies_threshold} factors."
+            f"Auto-strategies enabled. Will trigger after {auto_strategies_threshold} factors.",
        )
    else:
        quant_loop._auto_strategies = False
@@ -16,55 +16,26 @@ from rdagent.app.rl.ui.components import render_session, render_summary
 from rdagent.app.rl.ui.config import ALWAYS_VISIBLE_TYPES, OPTIONAL_TYPES
 from rdagent.app.rl.ui.data_loader import get_summary, get_valid_sessions, load_session
 from rdagent.app.rl.ui.rl_summary import render_job_summary
+from rdagent.core.utils import safe_resolve_path

 DEFAULT_LOG_BASE = "log/"


 def _safe_resolve(user_input: str | None, safe_root: Path) -> Path:
-    """
-    Resolve user path relative to safe_root; raise ValueError if it escapes.
-
-    Security: This function prevents path traversal attacks by:
-    1. Rejecting null bytes in user input
-    2. Rejecting Windows drive letters (C:\, D:\, etc.)
-    3. Rejecting absolute paths
-    4. Normalizing path to remove .. traversal attempts
-    5. Validating resolved path is within safe_root using a realpath-based check
-
-    All user-provided paths are validated before filesystem access.
-    """
-    # Treat the provided safe_root as trusted and canonicalize it once.
    safe_root = safe_root.expanduser().resolve()
-
-    # Empty input maps to the safe root directory.
    if not user_input:
        return safe_root
-
-    # Security check 1: Reject null bytes (path truncation attack)
    if "\x00" in user_input:
        raise ValueError("Invalid path: contains null byte")
-
    try:
-        # Security check 2: Normalize path to resolve .. and . components
        normalized = os.path.normpath(user_input.strip())
-
-        # Security check 3: Reject Windows drive letters (C:\, D:\, etc.)
        drive, _ = os.path.splitdrive(normalized)
        if drive:
            raise ValueError("Absolute paths with drive letters are not allowed")
-
-        # Security check 4: Reject absolute paths (/, //server/share, etc.)
        if os.path.isabs(normalized):
            raise ValueError("Absolute paths are not allowed")
-
-        # Security check 5: Build candidate path under safe_root and fully resolve it.
-        joined = os.path.join(str(safe_root), normalized)
-        resolved_candidate = os.path.realpath(joined)
-
-        # Security check 6: Validate candidate is within safe_root (prevent path traversal)
-        candidate_path = Path(resolved_candidate)
-        # Reconstruct from trusted safe_root so the returned path is root-derived.
-        return safe_root / candidate_path.relative_to(safe_root)
+        joined = safe_root / normalized
+        return safe_resolve_path(joined, safe_root)
    except (OSError, ValueError) as exc:
        raise ValueError(f"Invalid path outside of allowed root: {user_input}") from exc

@@ -82,7 +53,7 @@ def get_job_options(base_path: Path, safe_root: Path | None = None) -> list[str]

    # Security fix: Validate base_path to prevent path traversal
    try:
-        base_path_resolved = base_path.expanduser().resolve()
+        base_path_resolved = base_path.expanduser().resolve()  # nosec B614 — validated against safe_root below via relative_to()

        if safe_root is not None:
            safe_root_resolved = safe_root.expanduser().resolve()
@@ -203,8 +174,7 @@ def main():
        except ValueError as e:
            st.warning(str(e))
            return
-        # job_path is validated by _safe_resolve() above
-        if job_path.exists():  # nosec B614 – path validated by _safe_resolve
+        if job_path.exists():
            render_job_summary(job_path, safe_root, is_root=is_root_job)
        else:
            st.warning(f"Job folder not found: {job_folder}")
@@ -15,6 +15,7 @@ from typing import Any
 import streamlit as st

 from rdagent.app.rl.ui.config import EventType
+from rdagent.core.utils import safe_resolve_path
 from rdagent.log.storage import FileStorage


@@ -76,11 +77,10 @@ def extract_stage(tag: str) -> str:
 def get_valid_sessions(log_folder: Path, safe_root: Path | None = None) -> list[str]:
    """Get list of valid session directories, optionally validating against a safe root."""
    if safe_root is not None:
-        root_real = os.path.realpath(str(safe_root.expanduser()))
-        folder_real = os.path.realpath(str(log_folder.expanduser()))
-        if not (folder_real == root_real or folder_real.startswith(root_real + os.sep)):
+        try:
+            log_folder = safe_resolve_path(log_folder, safe_root)
+        except ValueError:
            return []
-        log_folder = Path(folder_real)

    if not log_folder.exists():
        return []
@@ -245,13 +245,11 @@ def parse_event(tag: str, content: Any, timestamp: datetime) -> Event | None:
@st.cache_data(ttl=300, hash_funcs={Path: str})
 def load_session(log_path: Path, safe_root: Path | None = None) -> Session:
    """Load events into hierarchical session structure, optionally validating against safe root."""
-    # Validate path is within safe_root if provided
    if safe_root is not None:
-        root_real = os.path.realpath(str(safe_root.expanduser()))
-        path_real = os.path.realpath(str(log_path.expanduser()))
-        if not (path_real == root_real or path_real.startswith(root_real + os.sep)):
+        try:
+            log_path = safe_resolve_path(log_path, safe_root)
+        except ValueError:
            return Session()
-        log_path = Path(path_real)

    session = Session()

@@ -9,6 +9,8 @@ from pathlib import Path
 import pandas as pd
 import streamlit as st

+from rdagent.core.utils import safe_resolve_path
+

 def is_valid_task(task_path: Path) -> bool:
    """Check if directory is a valid RL task (has __session__ subdirectory)"""
@@ -62,14 +64,10 @@ def get_loop_status(task_path: Path, loop_id: int) -> tuple[str, bool | None]:


 def _validate_job_path(job_path: Path, safe_root: Path) -> Path:
-    """Resolve and validate that job_path stays within safe_root."""
-    resolved_root = safe_root.expanduser().resolve()
-    resolved_job = job_path.expanduser().resolve()
    try:
-        # Reconstruct from trusted root so the returned path is root-derived.
-        return resolved_root / resolved_job.relative_to(resolved_root)
+        return safe_resolve_path(job_path, safe_root)
    except ValueError:
-        raise ValueError(f"Job path is outside allowed root {resolved_root}")
+        raise ValueError(f"Job path is outside allowed root {safe_root}")


 def get_max_loops(job_path: Path, safe_root: Path | None = None) -> int:
@@ -54,11 +54,11 @@ def rdagent_info():
    current_version = importlib.metadata.version("rdagent")
    logger.info(f"RD-Agent version: {current_version}")
    api_url = f"https://api.github.com/repos/microsoft/RD-Agent/contents/requirements.txt?ref=main"
-    response = requests.get(api_url)
+    response = requests.get(api_url, timeout=30)
    if response.status_code == 200:
        files = response.json()
        file_url = files["download_url"]
-        file_response = requests.get(file_url)
+        file_response = requests.get(file_url, timeout=30)
        if file_response.status_code == 200:
            all_file_contents = file_response.text.split("\n")
        else:
@@ -5,13 +5,29 @@ from .risk_management import CorrelationAnalyzer, PortfolioOptimizer, AdvancedRi
 from .vbt_backtest import (
    DEFAULT_BARS_PER_YEAR,
    DEFAULT_TXN_COST_BPS,
+    FTMO_INITIAL_CAPITAL,
+    FTMO_MAX_DAILY_LOSS,
+    FTMO_MAX_TOTAL_LOSS,
+    FTMO_MAX_LEVERAGE,
+    FTMO_RISK_PER_TRADE,
+    OOS_START_DEFAULT,
+    WF_IS_YEARS,
+    WF_OOS_YEARS,
+    WF_STEP_YEARS,
    backtest_from_forward_returns,
    backtest_signal,
+    backtest_signal_ftmo,
+    monte_carlo_trade_pvalue,
+    walk_forward_rolling,
 )

 __all__ = [
    'BacktestMetrics', 'FactorBacktester', 'ResultsDatabase',
    'CorrelationAnalyzer', 'PortfolioOptimizer', 'AdvancedRiskManager',
-    'backtest_signal', 'backtest_from_forward_returns',
+    'backtest_signal', 'backtest_signal_ftmo', 'backtest_from_forward_returns',
+    'monte_carlo_trade_pvalue', 'walk_forward_rolling',
    'DEFAULT_BARS_PER_YEAR', 'DEFAULT_TXN_COST_BPS',
+    'FTMO_INITIAL_CAPITAL', 'FTMO_MAX_DAILY_LOSS', 'FTMO_MAX_TOTAL_LOSS',
+    'FTMO_MAX_LEVERAGE', 'FTMO_RISK_PER_TRADE', 'OOS_START_DEFAULT',
+    'WF_IS_YEARS', 'WF_OOS_YEARS', 'WF_STEP_YEARS',
 ]
@@ -72,7 +72,7 @@ class BacktestMetrics:
 class FactorBacktester:
    def __init__(self):
        self.metrics = BacktestMetrics()
-        self.results_path = Path(__file__).parent.parent.parent / "results" / "backtests"
+        self.results_path = Path(__file__).parent.parent.parent.parent / "results" / "backtests"
        self.results_path.mkdir(parents=True, exist_ok=True)

    def run_backtest(
@@ -222,7 +222,7 @@ class FactorBacktester:

            # Calculate return for this step
            if step > 0:
-                prev_price = float(price_values[step - 1]) if step > 0 else current_price
+                prev_price = float(price_values[step - 1])
                if prev_price > 0:
                    step_return = (current_price - prev_price) / prev_price * position
                    returns_history.append(step_return)
@@ -71,6 +71,9 @@ class ResultsDatabase:

        self.conn.commit()

+    _ALLOWED_TABLES = frozenset({"factors", "backtest_runs", "loop_results"})
+    _ALLOWED_COL_TYPES = frozenset({"REAL", "TEXT", "INTEGER", "BLOB"})
+
    def _add_column_if_not_exists(self, table: str, column: str, col_type: str) -> None:
        """
        Add a column to a table if it doesn't already exist.
@@ -78,20 +81,24 @@ class ResultsDatabase:
        Parameters
        ----------
        table : str
-            Table name
+            Table name (must be in _ALLOWED_TABLES)
        column : str
-            Column name to add
+            Column name to add (alphanumeric + underscore only)
        col_type : str
-            SQL column type (e.g., 'REAL', 'TEXT')
+            SQL column type (must be in _ALLOWED_COL_TYPES)
        """
+        if table not in self._ALLOWED_TABLES:
+            raise ValueError(f"Unknown table: {table!r}")
+        if not column.replace("_", "").isalnum():
+            raise ValueError(f"Invalid column name: {column!r}")
+        if col_type not in self._ALLOWED_COL_TYPES:
+            raise ValueError(f"Invalid column type: {col_type!r}")
+
        c = self.conn.cursor()
-        try:
-            # Try to query the column - if it fails, it doesn't exist
-            # nosec B608: Internal schema migration, column names are controlled
-            c.execute(f"SELECT {column} FROM {table} LIMIT 1")  # nosec B608
-        except sqlite3.OperationalError:
-            # Column doesn't exist, add it
-            c.execute(f"ALTER TABLE {table} ADD COLUMN {column} {col_type}")  # nosec B608
+        c.execute("SELECT name FROM pragma_table_info(?)", (table,))
+        existing = {row[0] for row in c.fetchall()}
+        if column not in existing:
+            c.execute(f"ALTER TABLE {table} ADD COLUMN {column} {col_type}")
    
    def add_factor(self, name: str, type: str = "unknown") -> int:
        c = self.conn.cursor()
@@ -159,7 +166,7 @@ class ResultsDatabase:
        self.conn.commit()
        return c.lastrowid
    
-    def add_loop(self, loop_idx: int, success: int, fail: int, best_ic: float = None, status: str = "completed") -> int:
+    def add_loop(self, loop_idx: int, success: int, fail: int, best_ic: float | None = None, status: str = "completed") -> int:
        c = self.conn.cursor()
        rate = success / (success + fail) if (success + fail) > 0 else 0
        c.execute("""INSERT INTO loop_results (loop_index, factors_success, factors_fail, success_rate, best_ic, status)
@@ -183,16 +190,18 @@ class ResultsDatabase:
        pd.DataFrame
            DataFrame with factor names and metrics
        """
-        # Map shorthand to full column name
+        _ALLOWED_METRICS = frozenset({
+            'sharpe', 'ic', 'annual_return', 'max_drawdown',
+            'win_rate', 'information_ratio', 'volatility',
+        })
        metric_map = {
-            'sharpe': 'sharpe',
-            'ic': 'ic',
-            'return': 'annual_return',
-            'drawdown': 'max_drawdown',
-            'win_rate': 'win_rate',
+            'sharpe': 'sharpe', 'ic': 'ic', 'return': 'annual_return',
+            'drawdown': 'max_drawdown', 'win_rate': 'win_rate',
            'information_ratio': 'information_ratio',
        }
        col = metric_map.get(metric, metric)
+        if col not in _ALLOWED_METRICS:
+            raise ValueError(f"Unknown metric: {metric!r}")

        return pd.read_sql_query(
            f"""SELECT factor_name, ic, sharpe, annual_return, max_drawdown,
@@ -201,7 +210,7 @@ class ResultsDatabase:
                JOIN factors ON factor_id = factors.id
                WHERE {col} IS NOT NULL
                ORDER BY {col} DESC
-                LIMIT ?""",
+                LIMIT ?""",  # nosec B608 — col is validated against _ALLOWED_METRICS above
            self.conn,
            params=[limit]
        )
@@ -321,13 +330,13 @@ class ResultsDatabase:
        worst_drawdown = all_results['max_drawdown'].min() if total_runs > 0 and all_results['max_drawdown'].notna().any() else None

        # Scan factors directory for JSON files
-        factors_dir = Path(__file__).parent.parent.parent / "results" / "factors"
+        factors_dir = Path(__file__).parent.parent.parent.parent / "results" / "factors"
        json_factor_files = 0
        if factors_dir.exists():
            json_factor_files = len(list(factors_dir.glob("*.json")))

        # Scan failed runs
-        failed_dir = Path(__file__).parent.parent.parent / "results" / "failed_runs"
+        failed_dir = Path(__file__).parent.parent.parent.parent / "results" / "failed_runs"
        failed_runs_file = failed_dir / "failed_runs.json"
        failed_runs_count = 0
        failed_runs_data = []
@@ -1,21 +1,19 @@
 """
 Predix Risk Management - Korrelation, Portfolio-Optimierung
 """
+
 import numpy as np
 import pandas as pd
-from pathlib import Path
-from typing import Dict, List, Optional
-from datetime import datetime
-import json
+

 class CorrelationAnalyzer:
    def __init__(self, lookback: int = 60):
        self.lookback = lookback
-    
+
    def calculate_matrix(self, returns: pd.DataFrame) -> pd.DataFrame:
        return returns.dropna().corr()
-    
-    def find_uncorrelated(self, corr: pd.DataFrame, threshold: float = 0.3) -> List[str]:
+
+    def find_uncorrelated(self, corr: pd.DataFrame, threshold: float = 0.3) -> list[str]:
        result = []
        for f in corr.columns:
            others = [x for x in corr.columns if x != f]
@@ -28,9 +26,9 @@ class PortfolioOptimizer:
        try:
            w = np.linalg.inv(cov.values) @ exp_ret.values
            return w / np.sum(w)
-        except:
+        except (np.linalg.LinAlgError, ValueError):
            return np.ones(len(exp_ret)) / len(exp_ret)
-    
+
    def risk_parity(self, cov: pd.DataFrame, max_iter: int = 100) -> np.ndarray:
        n = cov.shape[0]
        w = np.ones(n) / n
@@ -53,36 +51,36 @@ class AdvancedRiskManager:
        self.max_dd = max_dd
        self.corr_analyzer = CorrelationAnalyzer()
        self.optimizer = PortfolioOptimizer()
-    
-    def check_limits(self, weights: np.ndarray, vol: float, dd: float) -> Dict[str, bool]:
+
+    def check_limits(self, weights: np.ndarray, vol: float, dd: float) -> dict[str, bool]:
        return {
-            'position_limit': np.max(np.abs(weights)) <= self.max_pos,
-            'leverage_limit': np.sum(np.abs(weights)) <= self.max_lev,
-            'drawdown_limit': abs(dd) <= self.max_dd,
+            "position_limit": np.max(np.abs(weights)) <= self.max_pos,
+            "leverage_limit": np.sum(np.abs(weights)) <= self.max_lev,
+            "drawdown_limit": abs(dd) <= self.max_dd,
        }

 if __name__ == "__main__":
    print("=== Risk Test ===")
    np.random.seed(42)
-    n, names = 252, ['Mom', 'MeanRev', 'Vol', 'Volu', 'ML']
+    n, names = 252, ["Mom", "MeanRev", "Vol", "Volu", "ML"]
    ret = pd.DataFrame(np.random.randn(n, 5), columns=names)
-    
+
    corr = CorrelationAnalyzer().calculate_matrix(ret)
    print("Korrelationsmatrix:")
    print(corr.round(2))
-    
+
    opt = PortfolioOptimizer()
    exp_ret = pd.Series([0.1, 0.08, 0.06, 0.07, 0.12], index=names)
    cov = ret.cov() * 252
-    
+
    mv = opt.mean_variance(exp_ret, cov)
    print("\nMean-Variance:")
    for n, w in zip(names, mv): print(f"  {n}: {w:.2%}")
-    
+
    rp = opt.risk_parity(cov)
    print("\nRisk Parity:")
    for n, w in zip(names, rp): print(f"  {n}: {w:.2%}")
-    
+
    rm = AdvancedRiskManager()
    checks = rm.check_limits(mv, 0.15, -0.08)
    print(f"\nLimits OK: {all(checks.values())}")
@@ -19,7 +19,7 @@ Design goals
 """
 from __future__ import annotations

-from typing import Any, Dict, Optional
+from typing import Any

 import numpy as np
 import pandas as pd
@@ -32,10 +32,22 @@ except ImportError:
    VBT_AVAILABLE = False


-DEFAULT_TXN_COST_BPS = 1.5
+# 2.35 pip realistic EUR/USD cost: 1.5 spread + 0.5 slippage + 0.35 commission
+# At EUR/USD ≈ 1.10:  2.35 pip * (0.0001/1.10) ≈ 2.14 bps of notional.
+DEFAULT_TXN_COST_BPS = 2.14
 DEFAULT_BARS_PER_YEAR = 252 * 1440  # 252 trading days * 1440 min/day = 362,880
 EXTREME_BAR_THRESHOLD = 0.05  # |ret| > 5% on a single 1-min bar → suspicious

+# FTMO 100k account rules (enforced in backtest_signal when ftmo=True)
+FTMO_INITIAL_CAPITAL   = 100_000.0
+FTMO_MAX_DAILY_LOSS    = 0.05   # 5%  of initial → block new trades rest of day
+FTMO_MAX_TOTAL_LOSS    = 0.10   # 10% of initial → simulation ends
+# Risk-based position sizing: 0.5% equity risk per trade, 10-pip stop, max 1:30 leverage
+FTMO_RISK_PER_TRADE    = 0.005
+FTMO_STOP_PIPS         = 10
+FTMO_PIP               = 0.0001
+FTMO_MAX_LEVERAGE      = 30
+

 def _compute_trade_pnl(position: pd.Series, strategy_returns: pd.Series) -> pd.Series:
    """
@@ -55,9 +67,8 @@ def _cross_check_with_vbt(
    close: pd.Series,
    position: pd.Series,
    txn_cost: float,
-    manual_total_return: float,
    freq: str,
-) -> Optional[float]:
+) -> float | None:
    """Run a vectorbt simulation and return its total_return for comparison."""
    if not VBT_AVAILABLE:
        return None
@@ -72,7 +83,8 @@ def _cross_check_with_vbt(
            init_cash=10_000.0,
            freq=freq,
        )
-        return float(pf.total_return())
+        tr = float(pf.total_return())
+        return tr if np.isfinite(tr) else None
    except Exception:
        return None

@@ -83,9 +95,9 @@ def backtest_signal(
    txn_cost_bps: float = DEFAULT_TXN_COST_BPS,
    freq: str = "1min",
    bars_per_year: int = DEFAULT_BARS_PER_YEAR,
-    forward_returns: Optional[pd.Series] = None,
+    forward_returns: pd.Series | None = None,
    cross_check: bool = False,
-) -> Dict[str, Any]:
+) -> dict[str, Any]:
    """
    Run a single-asset backtest from a position signal.

@@ -192,7 +204,7 @@ def backtest_signal(
    calmar = ann_return_arith / abs(max_dd) if max_dd < 0 else 0.0

    trade_pnl = _compute_trade_pnl(position, strategy_returns)
-    n_trades = int(len(trade_pnl))
+    n_trades = len(trade_pnl)
    n_position_changes = int((position.diff().fillna(0) != 0).sum())

    if n_trades > 0:
@@ -204,7 +216,7 @@ def backtest_signal(
        win_rate = 0.0
        profit_factor = 0.0

-    ic: Optional[float] = None
+    ic: float | None = None
    if forward_returns is not None:
        fwd = pd.to_numeric(forward_returns, errors="coerce")
        common = signal.index.intersection(fwd.dropna().index)
@@ -215,7 +227,7 @@ def backtest_signal(
                ic_val = float(s.corr(f))
                ic = ic_val if np.isfinite(ic_val) else None

-    result: Dict[str, Any] = {
+    result: dict[str, Any] = {
        "status": "success",
        "sharpe": sharpe,
        "sortino": sortino,
@@ -232,7 +244,7 @@ def backtest_signal(
        "volatility": volatility,
        "n_trades": n_trades,
        "n_position_changes": n_position_changes,
-        "n_bars": int(len(strategy_returns)),
+        "n_bars": len(strategy_returns),
        "n_months": float(n_months),
        "signal_long": int((signal > 0).sum()),
        "signal_short": int((signal < 0).sum()),
@@ -252,19 +264,341 @@ def backtest_signal(
            close=close,
            position=position,
            txn_cost=txn_cost,
-            manual_total_return=total_return,
            freq=freq,
        )

    return result


+def _apply_ftmo_mask(
+    signal: pd.Series,
+    close: pd.Series,
+    leverage: float,
+    txn_cost_bps: float,
+) -> tuple[pd.Series, dict]:
+    """
+    Apply FTMO daily/total loss rules to a signal series.
+
+    Returns a masked signal (positions zeroed after each limit breach) and
+    a dict of FTMO compliance metrics.
+    """
+    txn_cost = txn_cost_bps / 10_000.0
+    position = signal.shift(1).fillna(0) * leverage
+    bar_ret  = close.pct_change().fillna(0)
+
+    equity   = FTMO_INITIAL_CAPITAL
+    peak_day = FTMO_INITIAL_CAPITAL
+    masked   = signal.copy()
+
+    daily_breaches = 0
+    total_breached = False
+    total_breach_ts: pd.Timestamp | None = None
+    current_day    = None
+    day_start_eq   = FTMO_INITIAL_CAPITAL
+
+    pos_prev = 0.0
+    for ts, sig_i in signal.items():
+        day = ts.date() if hasattr(ts, "date") else ts
+
+        if day != current_day:
+            current_day  = day
+            day_start_eq = equity
+
+        pos_i  = float(signal.at[ts]) * leverage
+        ret_i  = float(bar_ret.get(ts, 0.0))
+        cost_i = abs(pos_i - pos_prev) * txn_cost
+        ret_frac = pos_prev * ret_i - cost_i
+        equity *= 1.0 + ret_frac if equity > 0 else 1.0
+        pos_prev = pos_i
+
+        if total_breached:
+            masked.at[ts] = 0
+            continue
+
+        daily_loss = (equity - day_start_eq) / FTMO_INITIAL_CAPITAL
+        total_loss = (equity - FTMO_INITIAL_CAPITAL) / FTMO_INITIAL_CAPITAL
+
+        if daily_loss < -FTMO_MAX_DAILY_LOSS:
+            daily_breaches += 1
+            day_start_eq = -999  # block rest of day
+            masked.at[ts] = 0
+
+        if total_loss < -FTMO_MAX_TOTAL_LOSS:
+            total_breached = True
+            total_breach_ts = ts
+            masked.at[ts] = 0
+
+    return masked, {
+        "ftmo_daily_breaches": daily_breaches,
+        "ftmo_total_breached": total_breached,
+        "ftmo_total_breach_ts": str(total_breach_ts) if total_breach_ts else None,
+        "ftmo_compliant": not total_breached and daily_breaches == 0,
+    }
+
+
+OOS_START_DEFAULT = "2024-01-01"
+
+# Rolling walk-forward default windows (IS years, OOS years, step years)
+WF_IS_YEARS  = 3
+WF_OOS_YEARS = 1
+WF_STEP_YEARS = 1
+
+
+def monte_carlo_trade_pvalue(
+    trade_pnl: pd.Series,
+    n_permutations: int = 1000,
+    seed: int = 0,
+) -> float:
+    """
+    Monte Carlo permutation test on trade-level P&L.
+
+    Runs a one-sided binomial test on trade-level win rate.
+
+    Tests H0: win_rate = 0.5 (random trading) against H1: win_rate > 0.5.
+    The ``n_permutations`` parameter is kept for API compatibility but is unused.
+
+    p < 0.05 → win rate is significantly above 50%, indicating a genuine per-trade edge.
+
+    Parameters
+    ----------
+    trade_pnl : pd.Series
+        Per-trade net returns (output of ``_compute_trade_pnl``).
+    n_permutations : int
+        Number of random permutations (default 1000).
+    seed : int
+        RNG seed for reproducibility.
+
+    Returns
+    -------
+    float
+        p-value in [0, 1]. Lower is better.
+    """
+    if len(trade_pnl) < 2:
+        return 1.0
+    trades = trade_pnl.values.copy()
+    # Binomial test: is the win rate significantly above 50%?
+    # p = probability of observing >= n_wins out of n_trades under null (win_rate=0.5).
+    # Low p → strategy has a significant positive edge per trade.
+    from scipy.stats import binomtest
+    n_wins = int((trades > 0).sum())
+    n_total = len(trades)
+    result = binomtest(n_wins, n_total, p=0.5, alternative="greater")
+    return float(result.pvalue)
+
+
+def walk_forward_rolling(
+    close: pd.Series,
+    signal: pd.Series,
+    leverage: float,
+    txn_cost_bps: float = DEFAULT_TXN_COST_BPS,
+    bars_per_year: int = DEFAULT_BARS_PER_YEAR,
+    is_years: int = WF_IS_YEARS,
+    oos_years: int = WF_OOS_YEARS,
+    step_years: int = WF_STEP_YEARS,
+) -> dict[str, Any]:
+    """
+    Rolling walk-forward validation: multiple IS/OOS windows shifted by ``step_years``.
+
+    Each window runs an independent FTMO simulation on the IS and OOS slices.
+    Produces aggregate OOS statistics to measure cross-time consistency.
+
+    Returns
+    -------
+    dict with keys:
+        wf_n_windows, wf_oos_sharpe_mean, wf_oos_sharpe_std,
+        wf_oos_monthly_return_mean, wf_oos_consistency (fraction of windows
+        with OOS Sharpe > 0), wf_windows (list of per-window dicts)
+    """
+    if not isinstance(close.index, pd.DatetimeIndex):
+        return {"wf_n_windows": 0}
+
+    start_year = close.index[0].year
+    end_year   = close.index[-1].year
+
+    windows = []
+    yr = start_year
+    while True:
+        is_start  = pd.Timestamp(f"{yr}-01-01")
+        is_end    = pd.Timestamp(f"{yr + is_years}-01-01")
+        oos_end   = pd.Timestamp(f"{yr + is_years + oos_years}-01-01")
+        if oos_end.year > end_year + 1:
+            break
+        is_mask  = (close.index >= is_start) & (close.index < is_end)
+        oos_mask = (close.index >= is_end)   & (close.index < oos_end)
+        if is_mask.sum() < 1000 or oos_mask.sum() < 1000:
+            yr += step_years
+            continue
+
+        window: dict[str, Any] = {
+            "is_start":  str(is_start.date()),
+            "is_end":    str(is_end.date()),
+            "oos_start": str(is_end.date()),
+            "oos_end":   str(oos_end.date()),
+        }
+        for mask, prefix in [(is_mask, "is"), (oos_mask, "oos")]:
+            close_s  = close.loc[mask]
+            signal_s = signal.loc[mask]
+            masked_s, _ = _apply_ftmo_mask(signal_s, close_s, leverage, txn_cost_bps)
+            r = backtest_signal(close=close_s, signal=masked_s,
+                                txn_cost_bps=txn_cost_bps, bars_per_year=bars_per_year)
+            window[f"{prefix}_sharpe"]       = r.get("sharpe", 0.0)
+            window[f"{prefix}_monthly_return_pct"] = r.get("monthly_return_pct", 0.0)
+            window[f"{prefix}_n_trades"]     = r.get("n_trades", 0)
+        windows.append(window)
+        yr += step_years
+
+    if not windows:
+        return {"wf_n_windows": 0}
+
+    oos_sharpes  = [w["oos_sharpe"] for w in windows]
+    oos_monthly  = [w["oos_monthly_return_pct"] for w in windows]
+    return {
+        "wf_n_windows":              len(windows),
+        "wf_oos_sharpe_mean":        float(np.mean(oos_sharpes)),
+        "wf_oos_sharpe_std":         float(np.std(oos_sharpes)),
+        "wf_oos_monthly_return_mean": float(np.mean(oos_monthly)),
+        "wf_oos_consistency":        float(np.mean([s > 0 for s in oos_sharpes])),
+        "wf_windows":                windows,
+    }
+
+
+def backtest_signal_ftmo(
+    close: pd.Series,
+    signal: pd.Series,
+    txn_cost_bps: float = DEFAULT_TXN_COST_BPS,
+    eurusd_price: float = 1.10,
+    risk_pct: float = FTMO_RISK_PER_TRADE,
+    stop_pips: float = FTMO_STOP_PIPS,
+    max_leverage: float = FTMO_MAX_LEVERAGE,
+    bars_per_year: int = DEFAULT_BARS_PER_YEAR,
+    forward_returns: pd.Series | None = None,
+    oos_start: str | None = OOS_START_DEFAULT,
+    wf_rolling: bool = False,
+    mc_n_permutations: int = 0,
+) -> dict[str, Any]:
+    """
+    FTMO-compliant backtest of a strategy signal on EUR/USD.
+
+    Applies on top of ``backtest_signal``:
+      - Realistic costs: default 2.14 bps (≈ 2.35 pip spread+slippage+commission)
+      - Risk-based position sizing: risk_pct equity per trade, stop_pips hard stop
+      - Max leverage cap: max_leverage (default 1:30, FTMO standard)
+      - FTMO daily loss limit (5%): positions zeroed rest of day after breach
+      - FTMO total loss limit (10%): all positions zeroed after breach
+      - FTMO-specific metrics added to result dict
+      - Walk-forward OOS split: IS metrics (before oos_start) + OOS metrics (after)
+
+    Parameters
+    ----------
+    close : pd.Series
+        1-min EUR/USD close prices.
+    signal : pd.Series
+        Raw strategy signal in {-1, 0, +1}.
+    txn_cost_bps : float
+        Transaction cost in bps (default 2.14 ≈ 2.35 pip on EUR/USD).
+    eurusd_price : float
+        Representative EUR/USD price for pip→bps conversion (default 1.10).
+    risk_pct : float
+        Fraction of equity risked per trade (default 0.005 = 0.5%).
+    stop_pips : float
+        Hard stop-loss distance in pips (default 10).
+    max_leverage : float
+        Maximum leverage (default 30 = FTMO 1:30).
+    oos_start : str or None
+        Start of out-of-sample period (ISO date). None disables OOS split.
+    wf_rolling : bool
+        If True, run rolling walk-forward validation (multiple IS/OOS windows).
+        Results are stored under ``wf_*`` keys. Default False.
+    mc_n_permutations : int
+        Number of Monte Carlo trade permutations. 0 = disabled (default).
+        When > 0, computes ``mc_pvalue``: fraction of permuted sequences whose
+        total return >= real total return. p < 0.05 indicates a genuine edge.
+    """
+    stop_price = stop_pips * FTMO_PIP
+    leverage_by_risk = risk_pct / (stop_price / eurusd_price)
+    leverage = min(leverage_by_risk, max_leverage)
+
+    masked_signal, ftmo_metrics = _apply_ftmo_mask(signal, close, leverage, txn_cost_bps)
+
+    result = backtest_signal(
+        close=close,
+        signal=masked_signal,
+        txn_cost_bps=txn_cost_bps,
+        bars_per_year=bars_per_year,
+        forward_returns=forward_returns,
+    )
+
+    result.update(ftmo_metrics)
+    result["ftmo_leverage"] = round(leverage, 2)
+    result["ftmo_risk_pct"] = risk_pct
+    result["ftmo_stop_pips"] = stop_pips
+
+    # Re-scale reported equity metrics to FTMO_INITIAL_CAPITAL
+    result["ftmo_end_equity"] = FTMO_INITIAL_CAPITAL * (1 + result.get("total_return", 0))
+    result["ftmo_monthly_profit"] = FTMO_INITIAL_CAPITAL * result.get("monthly_return", 0)
+
+    # Walk-forward OOS split
+    if oos_start is not None:
+        oos_ts = pd.Timestamp(oos_start)
+        is_mask  = close.index < oos_ts
+        oos_mask = close.index >= oos_ts
+
+        def _split_bt(mask: pd.Series[bool], prefix: str) -> None:
+            if mask.sum() < 100:
+                return
+            close_s  = close.loc[mask]
+            signal_s = signal.loc[mask]  # raw signal, not masked — fresh FTMO sim per period
+            fwd_split = forward_returns.loc[mask] if forward_returns is not None else None
+            masked_s, _ = _apply_ftmo_mask(signal_s, close_s, leverage, txn_cost_bps)
+            split_result = backtest_signal(
+                close=close_s,
+                signal=masked_s,
+                txn_cost_bps=txn_cost_bps,
+                bars_per_year=bars_per_year,
+                forward_returns=fwd_split,
+            )
+            for k, v in split_result.items():
+                if k not in ("equity_curve", "status"):
+                    result[f"{prefix}_{k}"] = v
+
+        _split_bt(is_mask,  "is")
+        _split_bt(oos_mask, "oos")
+
+        result["oos_start"] = oos_start
+        result["is_n_bars"]  = int(is_mask.sum())
+        result["oos_n_bars"] = int(oos_mask.sum())
+
+    # Rolling walk-forward validation
+    if wf_rolling:
+        wf = walk_forward_rolling(
+            close=close,
+            signal=signal,
+            leverage=leverage,
+            txn_cost_bps=txn_cost_bps,
+            bars_per_year=bars_per_year,
+        )
+        result.update(wf)
+
+    # Monte Carlo trade permutation test
+    if mc_n_permutations > 0:
+        position = masked_signal.shift(1).fillna(0)
+        bar_ret  = close.pct_change().fillna(0)
+        txn_cost = txn_cost_bps / 10_000.0
+        position_change = position.diff().abs().fillna(position.abs())
+        strat_ret = position * bar_ret - position_change * txn_cost
+        trade_pnl = _compute_trade_pnl(position, strat_ret)
+        result["mc_pvalue"]        = monte_carlo_trade_pvalue(trade_pnl, mc_n_permutations)
+        result["mc_n_permutations"] = mc_n_permutations
+
+    return result
+
+
 def backtest_from_forward_returns(
    factor_values: pd.Series,
    forward_returns: pd.Series,
    txn_cost_bps: float = DEFAULT_TXN_COST_BPS,
    bars_per_year: int = DEFAULT_BARS_PER_YEAR,
-) -> Dict[str, Any]:
+) -> dict[str, Any]:
    """
    Backtest a factor using sign(factor) as signal against forward returns.

@@ -302,7 +636,7 @@ def backtest_from_forward_returns(
    ic = ic_val if np.isfinite(ic_val) else 0.0

    trade_pnl = _compute_trade_pnl(position, strategy_returns)
-    n_trades = int(len(trade_pnl))
+    n_trades = len(trade_pnl)
    win_rate = float((trade_pnl > 0).mean()) if n_trades > 0 else 0.0

    ann_return = float(strategy_returns.mean() * bars_per_year)
@@ -318,7 +652,7 @@ def backtest_from_forward_returns(
        "win_rate": win_rate,
        "n_trades": n_trades,
        "ic": ic,
-        "n_bars": int(len(strategy_returns)),
+        "n_bars": len(strategy_returns),
        "txn_cost_bps": txn_cost_bps,
        "bars_per_year": bars_per_year,
    }
@@ -75,8 +75,10 @@ class CoSTEER(Developer[Experiment]):

    def _get_last_fb(self) -> CoSTEERMultiFeedback:
        fb = self.evolve_agent.evolving_trace[-1].feedback
-        assert fb is not None, "feedback is None"
-        assert isinstance(fb, CoSTEERMultiFeedback), "feedback must be of type CoSTEERMultiFeedback"
+        if fb is None:
+            raise AssertionError("feedback is None")
+        if not isinstance(fb, CoSTEERMultiFeedback):
+            raise TypeError("feedback must be of type CoSTEERMultiFeedback")
        return fb

    def should_use_new_evo(self, base_fb: CoSTEERMultiFeedback | None, new_fb: CoSTEERMultiFeedback) -> bool:
@@ -121,7 +123,8 @@ class CoSTEER(Developer[Experiment]):

        for evo_exp in self.evolve_agent.multistep_evolve(evo_exp, self.evaluator):
            iteration_count += 1
-            assert isinstance(evo_exp, Experiment)  # multiple inheritance
+            if not isinstance(evo_exp, Experiment):
+                raise TypeError("evo_exp must be an instance of Experiment")
            evo_fb = self._get_last_fb()
            update_fallback = self.should_use_new_evo(
                base_fb=fallback_evo_fb,
@@ -154,7 +157,8 @@ class CoSTEER(Developer[Experiment]):
                evo_exp = fallback_evo_exp
                evo_exp.recover_ws_ckp()
                evo_fb = fallback_evo_fb
-            assert evo_fb is not None  # multistep_evolve should run at least once
+            if evo_fb is None:
+                raise AssertionError("multistep_evolve should run at least once")
            evo_exp = self._exp_postprocess_by_feedback(evo_exp, evo_fb)
        except CoderError as e:
            e.caused_by_timeout = reached_max_seconds
@@ -264,9 +268,12 @@ class CoSTEER(Developer[Experiment]):
        - Raise Error if it failed to handle the develop task
        -
        """
-        assert isinstance(evo, Experiment)
-        assert isinstance(feedback, CoSTEERMultiFeedback)
-        assert len(evo.sub_workspace_list) == len(feedback)
+        if not isinstance(evo, Experiment):
+            raise TypeError("evo must be an instance of Experiment")
+        if not isinstance(feedback, CoSTEERMultiFeedback):
+            raise TypeError("feedback must be an instance of CoSTEERMultiFeedback")
+        if len(evo.sub_workspace_list) != len(feedback):
+            raise ValueError("Length of sub_workspace_list must match length of feedback")

        # FIXME: when whould the feedback be None?
        failed_feedbacks = [
@@ -122,7 +122,8 @@ class MultiProcessEvolvingStrategy(EvolvingStrategy):
        last_feedback = None
        if len(evolving_trace) > 0:
            last_feedback = evolving_trace[-1].feedback
-            assert isinstance(last_feedback, CoSTEERMultiFeedback)
+            if not isinstance(last_feedback, CoSTEERMultiFeedback):
+                raise TypeError("last_feedback must be of type CoSTEERMultiFeedback")

        # 1.找出需要evolve的task
        to_be_finished_task_index: list[int] = []
@@ -1028,7 +1028,8 @@ class CoSTEERKnowledgeBaseV2(EvolvingKnowledgeBase):

        """
        node_count = len(nodes)
-        assert node_count >= 2, "nodes length must >=2"
+        if node_count < 2:
+            raise ValueError("nodes length must >=2")
        intersection_node_list = []
        if output_intersection_origin:
            origin_list = []
@@ -54,7 +54,8 @@ def get_ds_env(
        ValueError: If the env_type is not recognized.
    """
    conf = DSCoderCoSTEERSettings()
-    assert conf_type in ["kaggle", "mlebench"], f"Unknown conf_type: {conf_type}"
+    if conf_type not in ["kaggle", "mlebench"]:
+        raise ValueError(f"Unknown conf_type: {conf_type}")

    if conf.env_type == "docker":
        env_conf = DSDockerConf() if conf_type == "kaggle" else MLEBDockerConf()
@@ -79,7 +80,8 @@ def get_clear_ws_cmd(stage: Literal["before_training", "before_inference"] = "be
    """
    Clean the files in workspace to a specific stage
    """
-    assert stage in ["before_training", "before_inference"], f"Unknown stage: {stage}"
+    if stage not in ["before_training", "before_inference"]:
+        raise ValueError(f"Unknown stage: {stage}")
    if DS_RD_SETTING.enable_model_dump and stage == "before_training":
        cmd = "rm -r submission.csv scores.csv models trace.log"
    else:
@@ -13,7 +13,7 @@ File structure

 from pathlib import Path

-from jinja2 import Environment, StrictUndefined
+from jinja2 import Environment, StrictUndefined, select_autoescape

 from rdagent.app.data_science.conf import DS_RD_SETTING
 from rdagent.components.coder.CoSTEER.evaluators import (
@@ -88,7 +88,7 @@ class EnsembleMultiProcessEvolvingStrategy(MultiProcessEvolvingStrategy):
            code_spec = workspace.file_dict["spec/ensemble.md"]
        else:
            test_code = (
-                Environment(undefined=StrictUndefined)
+                Environment(undefined=StrictUndefined, autoescape=select_autoescape())
                .from_string((DIRNAME / "eval_tests" / "ensemble_test.txt").read_text())
                .render(
                    model_names=[
@@ -2,7 +2,7 @@ import json
 import re
 from pathlib import Path

-from jinja2 import Environment, StrictUndefined
+from jinja2 import Environment, StrictUndefined, select_autoescape

 from rdagent.app.data_science.conf import DS_RD_SETTING
 from rdagent.components.coder.CoSTEER.evaluators import (
@@ -55,7 +55,7 @@ class EnsembleCoSTEEREvaluator(CoSTEEREvaluator):
        fname = "test/ensemble_test.txt"
        test_code = (DIRNAME / "eval_tests" / "ensemble_test.txt").read_text()
        test_code = (
-            Environment(undefined=StrictUndefined)
+            Environment(undefined=StrictUndefined, autoescape=select_autoescape())
            .from_string(test_code)
            .render(
                model_names=[
@@ -51,13 +51,23 @@ class FactorAutoFixer:
        self.fixes_applied = []
        fixed_code = code

-        # Apply fixes in order - groupby fixes MUST come before min_periods fixes
+        # Apply fixes in order
+        # NOTE: _fix_min_periods is intentionally excluded — it increased min_periods to
+        # match window size, which causes all-NaN output for intraday data with 96 bars/day
+        # (window=240 > 96 means zero valid bars per day). The LLM sets its own min_periods.
        fix_methods = [
-            self._fix_groupby_apply_to_transform,  # First: fix groupby patterns
-            self._fix_min_periods,                  # Second: fix min_periods in resulting rolling calls
-            self._fix_inf_nan_handling,             # Third: add inf/nan handling
-            self._fix_data_range_processing,        # Fourth: ensure full data range
-            self._fix_multiindex_groupby,           # Fifth: ensure groupby on MultiIndex
+            self._fix_instrument_column_access,     # First: fix df['instrument'] on MultiIndex
+            self._fix_instrument_loc_multiindex,    # Second: fix df.loc[instrument_var] on MultiIndex
+            self._fix_zero_volume_proxy,            # Third: replace zero $volume with range proxy
+            self._fix_reset_index_groupby,          # Fourth: fix groupby(level=N) after reset_index()
+            self._fix_groupby_mixed_levels,         # Fifth: fix groupby(level=[int, str])
+            self._fix_groupby_column_on_multiindex, # Sixth: fix groupby(['instrument','date']) on MultiIndex
+            self._fix_chained_groupby,              # Seventh: fix groupby(level=N).groupby('date') chain
+            self._fix_rolling_ddof,                 # Eighth: remove unsupported ddof kwarg
+            self._fix_groupby_apply_to_transform,   # Ninth: fix groupby patterns
+            self._fix_inf_nan_handling,             # Tenth: add inf/nan handling
+            self._fix_data_range_processing,        # Eleventh: ensure full data range
+            self._fix_multiindex_groupby,           # Twelfth: ensure groupby on MultiIndex
        ]

        for fix_method in fix_methods:
@@ -75,6 +85,352 @@ class FactorAutoFixer:

        return fixed_code

+    def _fix_instrument_column_access(self, code: str) -> str:
+        """
+        Fix: df['instrument'] raises KeyError on a MultiIndex DataFrame because
+        'instrument' is an index level (level 1), not a column.
+
+        Replace df['instrument'] with df.index.get_level_values('instrument')
+        but only when the DataFrame has a MultiIndex (not after reset_index which
+        would have promoted it to a real column).
+
+        Also fixes df.reset_index()['instrument'] correctly since after reset_index
+        the column exists.
+        """
+        fixed_code = code
+
+        # Skip if already fixed or if reset_index() is being used before the access
+        # We only fix bare df['instrument'] where df is the original MultiIndex frame.
+        # Heuristic: if the assignment lhs or context shows reset_index, leave it alone.
+
+        # Pattern: <varname>['instrument'] where varname is NOT a reset_index result
+        reset_vars = set(re.findall(r'(\w+)\s*=\s*\w[^=\n]*\.reset_index\(', fixed_code))
+
+        def _replace_instrument_access(m: re.Match) -> str:
+            var = m.group(1)
+            if var in reset_vars:
+                return m.group(0)  # leave reset_index vars alone — column exists
+            self.fixes_applied.append(f"instrument_column: {var}['instrument'] → get_level_values(1)")
+            return f"{var}.index.get_level_values(1)"
+
+        # Exclude assignment targets: var['instrument'] = ... must not become
+        # var.index.get_level_values(1) = ... (SyntaxError: cannot assign to function call)
+        fixed_code = re.sub(r"(\w+)\['instrument'\](?!\s*=)", _replace_instrument_access, fixed_code)
+
+        return fixed_code
+
+    def _fix_instrument_loc_multiindex(self, code: str) -> str:
+        """
+        Fix: df.loc[instrument_var] raises DateParseError on a (datetime, instrument)
+        MultiIndex because pandas tries to match the instrument string against the
+        datetime level (level 0).
+
+        Pattern detected: for-loops iterating over get_level_values('instrument') or
+        get_level_values(1) where the loop variable is then used as df.loc[loop_var].
+
+        Replacement: df.loc[instrument_var] → df.xs(instrument_var, level=1)
+        """
+        fixed_code = code
+
+        # Find variables iterated from get_level_values('instrument') or get_level_values(1)
+        inst_vars = set(
+            re.findall(
+                r"for\s+(\w+)\s+in\s+.+?\.get_level_values\s*\(\s*(?:1|['\"]instrument['\"])\s*\)[^:\n]*:",
+                code,
+            )
+        )
+
+        if not inst_vars:
+            return fixed_code
+
+        for var in inst_vars:
+            # Replace DF.loc[var] (read) with DF.xs(var, level=1)
+            # Exclude write-back patterns (DF.loc[var] = ...) — leave those as-is
+            def _make_replacer(v: str):
+                def _replace(m: re.Match) -> str:
+                    df_var = m.group(1)
+                    self.fixes_applied.append(
+                        f"instrument_loc: {df_var}.loc[{v}] → {df_var}.xs({v}, level=1)"
+                    )
+                    return f"{df_var}.xs({v}, level=1)"
+
+                return _replace
+
+            # Only match when NOT followed by ' =' (assignment)
+            fixed_code = re.sub(
+                rf"(\w+)\.loc\[\s*{re.escape(var)}\s*\](?!\s*=)",
+                _make_replacer(var),
+                fixed_code,
+            )
+
+        return fixed_code
+
+    def _fix_zero_volume_proxy(self, code: str) -> str:
+        """
+        Fix: $volume is always 0 in our EUR/USD dataset (FX has no real volume).
+        Any factor using $volume (VWAP, volume-weighted returns, etc.) produces
+        all-NaN output because 0*price=0 and sum(0)/sum(0)=NaN.
+
+        Insert a guard right after pd.read_hdf() that replaces zero volume with
+        the intraday price-range proxy ($high - $low) so volume-weighted factors
+        produce meaningful signals.
+        """
+        if "'$volume'" not in code and '"$volume"' not in code:
+            return code
+
+        # Already patched
+        if "volume proxy" in code:
+            return code
+
+        lines = code.splitlines()
+        insert_after = -1
+        df_var = "df"
+        indent = "    "
+
+        for i, line in enumerate(lines):
+            if "read_hdf(" in line:
+                m = re.match(r"(\s*)(\w+)\s*=\s*", line)
+                if m:
+                    indent = m.group(1)
+                    df_var = m.group(2)
+                else:
+                    m2 = re.match(r"(\s*)", line)
+                    indent = m2.group(1) if m2 else "    "
+                insert_after = i
+                break
+
+        if insert_after == -1:
+            return code
+
+        proxy_lines = [
+            f"{indent}# volume proxy: $volume is always 0 in FX data — use price-range as proxy",
+            f"{indent}if ({df_var}['$volume'] == 0).all():",
+            f"{indent}    {df_var}['$volume'] = {df_var}['$high'] - {df_var}['$low']",
+        ]
+        lines = lines[: insert_after + 1] + proxy_lines + lines[insert_after + 1 :]
+        self.fixes_applied.append("volume_proxy: replaced zero $volume with ($high - $low)")
+        return "\n".join(lines)
+
+    def _fix_reset_index_groupby(self, code: str) -> str:
+        """
+        Fix: groupby(level=N) on a variable created by .reset_index() fails because
+        reset_index() converts the MultiIndex into regular columns, leaving a plain
+        RangeIndex.  Replace groupby(level=N) on such variables with
+        groupby('instrument').
+
+        Detected pattern:
+            varname = <anything>.reset_index(...)
+            ...
+            varname.groupby(level=0|1)
+        """
+        fixed_code = code
+
+        # Find all variables assigned via reset_index()
+        reset_vars = set(re.findall(r'(\w+)\s*=\s*\w[^=\n]*\.reset_index\(', fixed_code))
+
+        for var in reset_vars:
+            # Replace var.groupby(level=N) with var.groupby('instrument')
+            pattern = rf'{re.escape(var)}\.groupby\(level\s*=\s*\d+\)'
+            if re.search(pattern, fixed_code):
+                fixed_code = re.sub(pattern, f"{var}.groupby('instrument')", fixed_code)
+                self.fixes_applied.append(f"reset_index_groupby: {var}.groupby(level=N) → groupby('instrument')")
+
+        return fixed_code
+
+    def _fix_groupby_mixed_levels(self, code: str) -> str:
+        """
+        Fix: groupby(level=[int, 'str']) raises AssertionError because string level
+        names don't exist on an unnamed MultiIndex.  Keep only integer levels.
+
+        Pattern:  .groupby(level=[0, 'date'])  →  .groupby(level=0)
+                  .groupby(level=[1, 'date'])  →  .groupby(level=1)
+        """
+        fixed_code = code
+
+        def _keep_int_levels(m):
+            inner = m.group(1)
+            ints = re.findall(r'\b(\d+)\b', inner)
+            if not ints:
+                return m.group(0)
+            replacement = f'.groupby(level={ints[0]})' if len(ints) == 1 else f'.groupby(level=[{", ".join(ints)}])'
+            self.fixes_applied.append(f"mixed_levels: groupby(level=[...,str]) → {replacement}")
+            return replacement
+
+        fixed_code = re.sub(r'\.groupby\(level=\[([^\]]+)\]\)', _keep_int_levels, fixed_code)
+        return fixed_code
+
+    def _fix_groupby_column_on_multiindex(self, code: str) -> str:
+        """
+        Fix: groupby(['instrument', 'date']) on a MultiIndex (datetime, instrument)
+        DataFrame fails with KeyError because those are index levels, not columns.
+
+        Correct replacement preserves BOTH dimensions so intraday calculations reset
+        per day:
+            var.groupby(['instrument', 'date'])
+            → var.groupby([var.index.get_level_values(1), var.index.get_level_values(0).normalize()])
+
+        Single-column groupby(['instrument']) is correctly replaced with groupby(level=1).
+        Note: do NOT convert groupby('instrument') → groupby(level=1) here — that would
+        undo the reset_index_groupby fix which correctly emits groupby('instrument').
+        """
+        fixed_code = code
+
+        # Variables created via reset_index() have a plain RangeIndex — applying
+        # get_level_values() on them would raise AttributeError. Skip those.
+        reset_vars = set(re.findall(r'(\w+)\s*=\s*\w[^=\n]*\.reset_index\(', fixed_code))
+
+        def _replace_two_col_groupby(m: re.Match, order: str) -> str:
+            var = m.group(1)
+            if var in reset_vars:
+                return m.group(0)  # leave reset_index vars alone — RangeIndex, not MultiIndex
+            if order == "instrument_date":
+                repl = (
+                    f"{var}.groupby([{var}.index.get_level_values(1), "
+                    f"{var}.index.get_level_values(0).normalize()])"
+                )
+            else:  # date_instrument
+                repl = (
+                    f"{var}.groupby([{var}.index.get_level_values(0).normalize(), "
+                    f"{var}.index.get_level_values(1)])"
+                )
+            self.fixes_applied.append(f"multiindex_groupby: {m.group(0)[:60]} → two-level")
+            return repl
+
+        # groupby(['instrument', 'date']) — capture variable name before .groupby
+        fixed_code = re.sub(
+            r'(\w+)\.groupby\(\[\'instrument\',\s*\'date\'\]\)',
+            lambda m: _replace_two_col_groupby(m, "instrument_date"),
+            fixed_code,
+        )
+        # groupby(['date', 'instrument'])
+        fixed_code = re.sub(
+            r'(\w+)\.groupby\(\[\'date\',\s*\'instrument\'\]\)',
+            lambda m: _replace_two_col_groupby(m, "date_instrument"),
+            fixed_code,
+        )
+        # single: groupby(['instrument']) → groupby(level=1), but not on reset_index vars
+        def _replace_single_instrument_groupby(m: re.Match) -> str:
+            # Look backwards to find the variable name
+            prefix = fixed_code[: m.start()]
+            var_match = re.search(r'(\w+)\s*$', prefix)
+            var = var_match.group(1) if var_match else ''
+            if var in reset_vars:
+                return m.group(0)
+            self.fixes_applied.append("multiindex_groupby: groupby(['instrument']) → groupby(level=1)")
+            return ".groupby(level=1)"
+
+        if re.search(r"\.groupby\(\['instrument'\]\)", fixed_code):
+            fixed_code = re.sub(r"\.groupby\(\['instrument'\]\)", _replace_single_instrument_groupby, fixed_code)
+
+        # groupby(level=['instrument', 'date']) — uses level= keyword with string names.
+        # 'date' is NOT a valid level name in our (datetime, instrument) MultiIndex;
+        # replace with get_level_values to normalize datetime to daily timestamps.
+        fixed_code = re.sub(
+            r"(\w+)\.groupby\(level=\['instrument',\s*'date'\]\)",
+            lambda m: (
+                self.fixes_applied.append(
+                    f"multiindex_groupby: {m.group(0)[:60]} → two-level get_level_values"
+                )
+                or f"{m.group(1)}.groupby([{m.group(1)}.index.get_level_values(1), "
+                f"{m.group(1)}.index.get_level_values(0).normalize()])"
+            ),
+            fixed_code,
+        )
+        # groupby(level=['date', 'instrument'])
+        fixed_code = re.sub(
+            r"(\w+)\.groupby\(level=\['date',\s*'instrument'\]\)",
+            lambda m: (
+                self.fixes_applied.append(
+                    f"multiindex_groupby: {m.group(0)[:60]} → two-level get_level_values"
+                )
+                or f"{m.group(1)}.groupby([{m.group(1)}.index.get_level_values(0).normalize(), "
+                f"{m.group(1)}.index.get_level_values(1)])"
+            ),
+            fixed_code,
+        )
+        # single: groupby(level=['instrument']) → groupby(level=1)
+        fixed_code = re.sub(
+            r"\.groupby\(level=\['instrument'\]\)",
+            lambda m: (self.fixes_applied.append("multiindex_groupby: groupby(level=['instrument']) → level=1") or ".groupby(level=1)"),
+            fixed_code,
+        )
+
+        return fixed_code
+
+    def _fix_chained_groupby(self, code: str) -> str:
+        """
+        Fix two broken patterns the LLM generates when trying to group by (instrument, date):
+
+        Pattern A — chained groupby (runtime AttributeError):
+            var.groupby(level=1).groupby('date')
+            → var.groupby([var.index.get_level_values(1),
+                           var.index.get_level_values(0).normalize()])
+
+        Pattern B — keyword arg inside list (SyntaxError):
+            var.groupby([level=1, 'date'])
+            → same two-level replacement
+        """
+        fixed_code = code
+
+        def _two_level(var: str, tag: str) -> str:
+            self.fixes_applied.append(f"chained_groupby: {tag} → two-level")
+            return (
+                f"{var}.groupby([{var}.index.get_level_values(1), "
+                f"{var}.index.get_level_values(0).normalize()])"
+            )
+
+        # Pattern A: var.groupby(level=N).groupby('date')
+        fixed_code = re.sub(
+            r'(\w+)\.groupby\(level=\d+\)\.groupby\(["\']date["\']\)',
+            lambda m: _two_level(m.group(1), m.group(0)[:60]),
+            fixed_code,
+        )
+
+        # Pattern B: .groupby([level=N, 'date']) — SyntaxError in Python.
+        # The variable before .groupby may be complex (e.g. df[mask]) so we don't
+        # try to capture it; we use df as the index reference (always correct since
+        # all filtered frames share df's MultiIndex structure).
+        def _two_level_df(tag: str) -> str:
+            self.fixes_applied.append(f"chained_groupby: {tag} → two-level")
+            return ".groupby([df.index.get_level_values(1), df.index.get_level_values(0).normalize()])"
+
+        fixed_code = re.sub(
+            r'\.groupby\(\[\s*level\s*=\s*\d+\s*,\s*["\']?date["\']?\s*\]\)',
+            lambda m: _two_level_df(m.group(0)[:60]),
+            fixed_code,
+        )
+        # Also handle reversed order: ['date', level=N]
+        fixed_code = re.sub(
+            r'\.groupby\(\[\s*["\']?date["\']?\s*,\s*level\s*=\s*\d+\s*\]\)',
+            lambda m: _two_level_df(m.group(0)[:60]),
+            fixed_code,
+        )
+
+        return fixed_code
+
+    def _fix_rolling_ddof(self, code: str) -> str:
+        """
+        Fix: pandas rolling() does not accept a ddof kwarg — raises TypeError.
+        Remove ddof from both rolling(..., ddof=N) and rolling(...).std(ddof=N).
+        """
+        fixed_code = code
+
+        # Form 1: ddof inside rolling() — .rolling(window=N, min_periods=M, ddof=K)
+        def _strip_ddof_from_rolling(m):
+            inner = re.sub(r',?\s*ddof\s*=\s*\d+', '', m.group(1))
+            inner = inner.strip(', ')
+            self.fixes_applied.append("rolling_ddof: removed ddof from rolling()")
+            return f'.rolling({inner})'
+
+        fixed_code = re.sub(r'\.rolling\(([^)]*ddof\s*=\s*\d+[^)]*)\)', _strip_ddof_from_rolling, fixed_code)
+
+        # Form 2: ddof inside .std() / .var() — .std(ddof=N)
+        if re.search(r'\.(std|var)\([^)]*ddof\s*=\s*\d+', fixed_code):
+            fixed_code = re.sub(r'\.(std|var)\([^)]*ddof\s*=\s*\d+[^)]*\)', r'.\1()', fixed_code)
+            self.fixes_applied.append("rolling_ddof: removed ddof from std()/var()")
+
+        return fixed_code
+
    def _fix_min_periods(self, code: str) -> str:
        """
        Fix: Ensure min_periods matches window size in rolling calculations.
@@ -325,6 +681,45 @@ class FactorAutoFixer:
            fixed_code = fixed_code.replace(old_code, new_code)
            self.fixes_applied.append(f"groupby: fixed rolling correlation (window={window}) with reset_index")

+        # === GENERAL FIX: DF.groupby(level=N)['col'].apply(lambda x: EXPR) ===
+        # apply() on a grouped Series returns a MultiIndex result (extra level prepended),
+        # causing index shape mismatch when assigned back to df['col'].
+        # Replace with transform() which preserves the original index.
+        col_apply_pattern = re.compile(
+            r"(\w+)\.groupby\(level=(\d+)\)\['([^']+)'\]\.apply\((\s*lambda\s+\w+\s*:.*?)\)",
+            re.DOTALL,
+        )
+        for m in list(col_apply_pattern.finditer(fixed_code)):
+            full = m.group(0)
+            df_var = m.group(1)
+            level = m.group(2)
+            col = m.group(3)
+            lam = m.group(4).strip()
+            new_expr = f"{df_var}.groupby(level={level})['{col}'].transform({lam})"
+            fixed_code = fixed_code.replace(full, new_expr, 1)
+            self.fixes_applied.append(
+                f"groupby: {df_var}.groupby(level={level})['{col}'].apply() → transform()"
+            )
+
+        # === FIX: .transform(...).reset_index(level=N, drop=True) ===
+        # transform() already returns the same index as the input — adding reset_index()
+        # after it drops an index level and causes ValueError on assignment back to df['col'].
+        # Detected line-by-line: if a line contains both .transform( and .reset_index(level=
+        reset_suffix = re.compile(r'\s*\.reset_index\s*\(\s*level\s*=[^,)]+,\s*drop\s*=\s*True\s*\)\s*$')
+        new_lines = []
+        changed = False
+        for line in fixed_code.splitlines():
+            if '.transform(' in line and '.reset_index(' in line:
+                cleaned = reset_suffix.sub('', line)
+                if cleaned != line:
+                    new_lines.append(cleaned)
+                    changed = True
+                    continue
+            new_lines.append(line)
+        if changed:
+            fixed_code = '\n'.join(new_lines)
+            self.fixes_applied.append("groupby: removed spurious .reset_index() after .transform()")
+
        # Pattern: Simple groupby().apply() with rolling().method()
        # df.groupby(level=N).apply(lambda x: x['col'].rolling(...).method())
        apply_pattern = r"df\.groupby\(level=(\d+)\)\.apply\(\s*lambda\s+x:\s+x\['([^']+)'\]\.rolling\([^)]+\)\.(\w+)\([^)]*\)\s*\)"
@@ -328,12 +328,13 @@ class FactorEqualValueRatioEvaluator(FactorEvaluator):
                "The source dataframe is None. Please check the implementation.",
                -1,
            )
+        acc_rate = -1
        try:
            close_values = gen_df.sub(gt_df).abs().lt(1e-6)
            result_int = close_values.astype(int)
            pos_num = result_int.sum().sum()
            acc_rate = pos_num / close_values.size
-        except:
+        except Exception:
            close_values = gen_df
        if close_values.all().iloc[0]:
            return (
@@ -161,8 +161,7 @@ class FactorFBWorkspace(FBWorkspace):

            try:
                subprocess.check_output(
-                    f"{FACTOR_COSTEER_SETTINGS.python_bin} {execution_code_path}",
-                    shell=True,
+                    [FACTOR_COSTEER_SETTINGS.python_bin, str(execution_code_path)],
                    cwd=self.workspace_path,
                    stderr=subprocess.STDOUT,
                    timeout=FACTOR_COSTEER_SETTINGS.file_based_execution_timeout,
@@ -53,7 +53,7 @@ evolving_strategy_factor_implementation_v1_system: |-
  - ALWAYS use `min_periods=N` where N equals the window size in rolling calculations (e.g., `.rolling(20, min_periods=20)`)
  - ALWAYS handle infinite values after division: `.replace([np.inf, -np.inf], np.nan)` before saving results
  - ALWAYS use `groupby(level=1)` or `groupby('instrument')` before rolling operations on MultiIndex dataframes
-  - Process the COMPLETE date range (2020-2026), do NOT filter by date
+  - Process the COMPLETE date range available in the HDF5 file (do NOT filter by date — the file may contain 2024 debug data or full 2020-2026 data)
  - Use `groupby().transform()` instead of `groupby().apply()` for single-column assignments

  Notice that you should not add any other text before or after the json format.
@@ -6,6 +6,7 @@ Two-step validation:
 2. Micro-batch testing - Runtime validation with small dataset
 """

+import ast
 import json
 import re
 import time
@@ -229,7 +230,7 @@ class LLMConfigValidator:
            final_metrics = re.search(r"\{'train_runtime':[^}]+\}", stdout)
            if final_metrics:
                try:
-                    metrics = eval(final_metrics.group(0))  # Safe: only numbers and strings
+                    metrics = ast.literal_eval(final_metrics.group(0))
                    result["final_metrics"] = {
                        "train_loss": metrics.get("train_loss"),
                        "train_runtime": metrics.get("train_runtime"),
@@ -123,8 +123,8 @@ model_cls = AntiSymmetricConv


 if __name__ == "__main__":
-    node_features = torch.load("node_features.pt")
-    edge_index = torch.load("edge_index.pt")
+    node_features = torch.load("node_features.pt", weights_only=True)
+    edge_index = torch.load("edge_index.pt", weights_only=True)

    # Model instantiation and forward pass
    model = AntiSymmetricConv(in_channels=node_features.size(-1))
@@ -78,8 +78,8 @@ model_cls = DirGNNConv


 if __name__ == "__main__":
-    node_features = torch.load("node_features.pt")
-    edge_index = torch.load("edge_index.pt")
+    node_features = torch.load("node_features.pt", weights_only=True)
+    edge_index = torch.load("edge_index.pt", weights_only=True)

    # Model instantiation and forward pass
    model = DirGNNConv(MessagePassing())
@@ -187,8 +187,8 @@ model_cls = GPSConv


 if __name__ == "__main__":
-    node_features = torch.load("node_features.pt")
-    edge_index = torch.load("edge_index.pt")
+    node_features = torch.load("node_features.pt", weights_only=True)
+    edge_index = torch.load("edge_index.pt", weights_only=True)

    # Model instantiation and forward pass
    model = GPSConv(channels=node_features.size(-1), conv=MessagePassing())
@@ -170,8 +170,8 @@ class LINKX(torch.nn.Module):
 model_cls = LINKX

 if __name__ == "__main__":
-    node_features = torch.load("node_features.pt")
-    edge_index = torch.load("edge_index.pt")
+    node_features = torch.load("node_features.pt", weights_only=True)
+    edge_index = torch.load("edge_index.pt", weights_only=True)

    # Model instantiation and forward pass
    model = LINKX(
@@ -102,8 +102,8 @@ class PMLP(torch.nn.Module):
 model_cls = PMLP

 if __name__ == "__main__":
-    node_features = torch.load("node_features.pt")
-    edge_index = torch.load("edge_index.pt")
+    node_features = torch.load("node_features.pt", weights_only=True)
+    edge_index = torch.load("edge_index.pt", weights_only=True)

    # Model instantiation and forward pass
    model = PMLP(
@@ -1180,8 +1180,8 @@ model_cls = ViSNet


 if __name__ == "__main__":
-    node_features = torch.load("node_features.pt")
-    edge_index = torch.load("edge_index.pt")
+    node_features = torch.load("node_features.pt", weights_only=True)
+    edge_index = torch.load("edge_index.pt", weights_only=True)

    # Model instantiation and forward pass
    model = ViSNet()
@@ -58,10 +58,12 @@ class ModelCodeEvaluator(CoSTEEREvaluator):
        model_execution_feedback: str = "",
        model_value_feedback: str = "",
    ):
-        assert isinstance(target_task, ModelTask)
-        assert isinstance(implementation, ModelFBWorkspace)
-        if gt_implementation is not None:
-            assert isinstance(gt_implementation, ModelFBWorkspace)
+        if not isinstance(target_task, ModelTask):
+            raise TypeError("target_task must be of type ModelTask")
+        if not isinstance(implementation, ModelFBWorkspace):
+            raise TypeError("implementation must be of type ModelFBWorkspace")
+        if gt_implementation is not None and not isinstance(gt_implementation, ModelFBWorkspace):
+            raise TypeError("gt_implementation must be of type ModelFBWorkspace")

        model_task_information = target_task.get_task_information()
        code = implementation.all_codes
@@ -113,10 +115,12 @@ class ModelFinalEvaluator(CoSTEEREvaluator):
        model_value_feedback: str,
        model_code_feedback: str,
    ):
-        assert isinstance(target_task, ModelTask)
-        assert isinstance(implementation, ModelFBWorkspace)
-        if gt_implementation is not None:
-            assert isinstance(gt_implementation, ModelFBWorkspace)
+        if not isinstance(target_task, ModelTask):
+            raise TypeError("target_task must be of type ModelTask")
+        if not isinstance(implementation, ModelFBWorkspace):
+            raise TypeError("implementation must be of type ModelFBWorkspace")
+        if gt_implementation is not None and not isinstance(gt_implementation, ModelFBWorkspace):
+            raise TypeError("gt_implementation must be of type ModelFBWorkspace")

        system_prompt = T(".prompts:evaluator_final_feedback.system").r(
            scenario=(
@@ -41,7 +41,8 @@ class ModelCoSTEEREvaluator(CoSTEEREvaluator):
                final_feedback="This task has failed too many times, skip implementation.",
                final_decision=False,
            )
-        assert isinstance(target_task, ModelTask)
+        if not isinstance(target_task, ModelTask):
+            raise TypeError(f"Expected ModelTask, got {type(target_task)}")

        # NOTE: Use fixed input to test the model to avoid randomness
        batch_size = 8
@@ -50,7 +51,8 @@ class ModelCoSTEEREvaluator(CoSTEEREvaluator):
        input_value = 0.4
        param_init_value = 0.6

-        assert isinstance(implementation, ModelFBWorkspace)
+        if not isinstance(implementation, ModelFBWorkspace):
+            raise TypeError(f"Expected ModelFBWorkspace, got {type(implementation)}")
        model_execution_feedback, gen_np_array = implementation.execute(
            batch_size=batch_size,
            num_features=num_features,
@@ -59,7 +61,8 @@ class ModelCoSTEEREvaluator(CoSTEEREvaluator):
            param_init_value=param_init_value,
        )
        if gt_implementation is not None:
-            assert isinstance(gt_implementation, ModelFBWorkspace)
+            if not isinstance(gt_implementation, ModelFBWorkspace):
+                raise TypeError(f"Expected ModelFBWorkspace, got {type(gt_implementation)}")
            _, gt_np_array = gt_implementation.execute(
                batch_size=batch_size,
                num_features=num_features,
@@ -125,8 +125,8 @@ class AntiSymmetricConv(torch.nn.Module):


 if __name__ == "__main__":
-    node_features = torch.load("node_features.pt")
-    edge_index = torch.load("edge_index.pt")
+    node_features = torch.load("node_features.pt", weights_only=True)
+    edge_index = torch.load("edge_index.pt", weights_only=True)

    # Model instantiation and forward pass
    model = AntiSymmetricConv(in_channels=node_features.size(-1))
@@ -23,7 +23,7 @@ import pandas as pd

 from rdagent.log import rdagent_logger as logger

-logger = logging.getLogger(__name__)
+_optuna_logger = logging.getLogger(__name__)

 try:
    import optuna
@@ -292,6 +292,7 @@ class OptunaOptimizer:
            "volatility_lookback": trial.suggest_int("volatility_lookback", 5, 500, step=5),
            "signal_bias": trial.suggest_float("signal_bias", -1.0, 1.0, step=0.05),
            "max_hold_bars": trial.suggest_int("max_hold_bars", 5, 1000, step=5),
+            "max_positions": trial.suggest_int("max_positions", 1, 5, step=1),
        }

    # Parameters that are allowed to be negative (not clamped to 0).
@@ -308,6 +309,7 @@ class OptunaOptimizer:
        "volatility_lookback": 1.0,
        "signal_bias": -1.0,
        "max_hold_bars": 1.0,
+        "max_positions": 1.0,
    }

    def _suggest_bounded(
@@ -357,6 +359,7 @@ class OptunaOptimizer:
            "volatility_lookback": (center.get("volatility_lookback", 100), 30),
            "signal_bias": (center.get("signal_bias", 0.0), 0.2),
            "max_hold_bars": (center.get("max_hold_bars", 100), 50),
+            "max_positions": (center.get("max_positions", 1), 2),
        }
        return {key: self._suggest_bounded(trial, key, c, hw) for key, (c, hw) in ranges.items()}

@@ -388,6 +391,7 @@ class OptunaOptimizer:
            "volatility_lookback": (center.get("volatility_lookback", 100), 10),
            "signal_bias": (center.get("signal_bias", 0.0), 0.07),
            "max_hold_bars": (center.get("max_hold_bars", 100), 17),
+            "max_positions": (center.get("max_positions", 1), 1),
        }
        return {key: self._suggest_bounded(trial, key, c, hw) for key, (c, hw) in ranges.items()}

@@ -467,6 +471,9 @@ class OptunaOptimizer:

            # Max holding periods (in bars)
            "max_hold_bars": trial.suggest_int("max_hold_bars", 10, 500, step=10),
+
+            # Max concurrent positions (1 = no pyramiding, 2-5 = scale-in)
+            "max_positions": trial.suggest_int("max_positions", 1, 5, step=1),
        }

        return params
@@ -597,6 +604,13 @@ class OptunaOptimizer:
            if signal_bias != 0.0:
                signal = (signal.astype(float) + signal_bias).round().astype(int).clip(-1, 1)

+            # Apply max_positions: scale signal by position_size_pct and cap exposure
+            max_positions   = int(params.get("max_positions", 1))
+            position_size_pct = float(params.get("position_size_pct", 1.0))
+            # Each "position" is position_size_pct of equity; total exposure capped at max_positions × size
+            effective_size  = min(position_size_pct * max_positions, 1.0)
+            signal = (signal.astype(float) * effective_size).clip(-1.0, 1.0)
+
            # Build a synthetic close from the factor-mean so we can route
            # through the same unified engine as every other backtest path.
            # Backtest formulas must match the orchestrator's real-OHLCV path.
@@ -605,16 +619,15 @@ class OptunaOptimizer:
            synthetic_close = (1 + combined_ret).cumprod() * 100.0

            from rdagent.components.backtesting.vbt_backtest import (
-                backtest_signal,
+                backtest_signal_ftmo,
                DEFAULT_TXN_COST_BPS,
            )
            import os as _os

-            bt = backtest_signal(
+            bt = backtest_signal_ftmo(
                close=synthetic_close,
                signal=signal,
                txn_cost_bps=float(_os.getenv("TXN_COST_BPS", DEFAULT_TXN_COST_BPS)),
-                freq="1min",
            )
            if bt.get("status") != "success":
                return self._default_metrics()
@@ -26,22 +26,18 @@ import traceback
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from datetime import datetime
 from pathlib import Path
-from typing import Any, Dict, List, Optional
+from typing import Any

 import numpy as np
 import pandas as pd
-import requests
-
-from rdagent.components.prompt_loader import load_prompt
 from rdagent.components.coder.optuna_optimizer import OptunaOptimizer
+from rdagent.components.prompt_loader import load_prompt

 # OHLCV data path
 OHLCV_PATH = Path(os.getenv(
-    'PREDIX_OHLCV_PATH',
-    '/home/nico/Predix/git_ignore_folder/factor_implementation_source_data/intraday_pv.h5'
+    "PREDIX_OHLCV_PATH",
+    "/home/nico/Predix/git_ignore_folder/factor_implementation_source_data/intraday_pv.h5",
 ))
-from rdagent.log import rdagent_logger as logger
-
 logger = logging.getLogger(__name__)


@@ -60,7 +56,7 @@ class StrategyOrchestrator:
        min_sharpe: float = 0.3,
        max_drawdown: float = -0.30,
        min_win_rate: float = 0.40,
-        results_dir: Optional[str] = None,
+        results_dir: str | None = None,
        use_optuna: bool = True,
        optuna_trials: int = 20,
        continuous_optimization: bool = True,
@@ -118,7 +114,7 @@ class StrategyOrchestrator:

        logger.info(
            f"StrategyOrchestrator initialized: style={self.trading_style}, "
-            f"top_factors={self.top_factors}, min_sharpe={self.min_sharpe}"
+            f"top_factors={self.top_factors}, min_sharpe={self.min_sharpe}",
        )

    def load_ohlcv_close(self) -> pd.Series:
@@ -126,31 +122,31 @@ class StrategyOrchestrator:
        if not OHLCV_PATH.exists():
            logger.warning(f"OHLCV data not found: {OHLCV_PATH}")
            return None
-        
+
        try:
-            ohlcv = pd.read_hdf(str(OHLCV_PATH), key='data')
-            if '$close' in ohlcv.columns:
-                close = ohlcv['$close'].dropna()
-            elif 'close' in ohlcv.columns:
-                close = ohlcv['close'].dropna()
+            ohlcv = pd.read_hdf(str(OHLCV_PATH), key="data")
+            if "$close" in ohlcv.columns:
+                close = ohlcv["$close"].dropna()
+            elif "close" in ohlcv.columns:
+                close = ohlcv["close"].dropna()
            else:
                close = ohlcv.select_dtypes(include=[np.number]).iloc[:, 0].dropna()
-            
+
            # Handle MultiIndex
            if isinstance(close.index, pd.MultiIndex):
                try:
-                    close = close.xs('EURUSD', level='instrument')
+                    close = close.xs("EURUSD", level="instrument")
                except KeyError:
-                    idx = close.index.get_level_values('instrument') == 'EURUSD'
+                    idx = close.index.get_level_values("instrument") == "EURUSD"
                    close = close[idx]
-                    close.index = close.index.droplevel('instrument')
-            
+                    close.index = close.index.droplevel("instrument")
+
            return close
        except Exception as e:
            logger.warning(f"Failed to load OHLCV data: {e}")
            return None

-    def load_top_factors(self) -> List[Dict[str, Any]]:
+    def load_top_factors(self) -> list[dict[str, Any]]:
        """
        Load top evaluated factors from JSON files.

@@ -177,7 +173,7 @@ class StrategyOrchestrator:

        # Sort by absolute IC and take top N
        factors.sort(key=lambda x: abs(x.get("ic", 0) or 0), reverse=True)
-        
+
        # Filter to only include factors that have parquet files
        factors_with_files = []
        for f in factors:
@@ -188,16 +184,16 @@ class StrategyOrchestrator:
                factors_with_files.append(f)
            else:
                logger.debug(f"Skipping {fname} - no parquet file")
-        
+
        # Select diverse factor TYPES, not just top IC
        # This ensures we get momentum, volatility, session, volume, etc.
        type_keywords = {
            "momentum": [], "trend": [], "volatility": [], "volume": [],
            "session": [], "london": [], "range": [], "vwap": [],
            "return": [], "ofi": [], "spread": [], "close": [],
-            "divergence": [], "other": []
+            "divergence": [], "other": [],
        }
-        
+
        for f in factors_with_files:
            name = f.get("factor_name", "").lower()
            matched = False
@@ -208,18 +204,18 @@ class StrategyOrchestrator:
                    break
            if not matched:
                type_keywords["other"].append(f)
-        
+
        # Select best from each type (ensures diversity)
        selected = []
        already_names = set()
-        
+
        # Priority order: momentum, divergence, volatility, session, volume, etc.
-        priority_types = ["momentum", "divergence", "volatility", "session", 
-                         "london", "range", "vwap", "volume", "ofi", "spread", 
+        priority_types = ["momentum", "divergence", "volatility", "session",
+                         "london", "range", "vwap", "volume", "ofi", "spread",
                         "return", "trend", "close", "other"]
-        
+
        per_type = max(2, self.top_factors // len(priority_types))
-        
+
        for kw in priority_types:
            for f in sorted(type_keywords[kw], key=lambda x: abs(x.get("ic", 0)), reverse=True):
                if f["factor_name"] not in already_names:
@@ -227,13 +223,13 @@ class StrategyOrchestrator:
                    already_names.add(f["factor_name"])
                    if len([s for s in selected if s["factor_name"] in [x["factor_name"] for x in type_keywords[kw]]]) >= per_type:
                        break
-        
+
        # Fill remaining with highest IC not yet selected
        if len(selected) < self.top_factors:
            remaining = [f for f in factors_with_files if f["factor_name"] not in already_names]
            remaining.sort(key=lambda x: abs(x.get("ic", 0)), reverse=True)
            selected.extend(remaining[:self.top_factors - len(selected)])
-        
+
        # Log diversity
        type_counts = {}
        for f in selected:
@@ -246,12 +242,12 @@ class StrategyOrchestrator:
                    break
            if not matched:
                type_counts["other"] = type_counts.get("other", 0) + 1
-        
+
        logger.info(f"Selected {len(selected)} diverse factors: {type_counts}")
-        
+
        return selected[:self.top_factors]

-    def load_factor_values(self, factor_name: str) -> Optional[pd.Series]:
+    def load_factor_values(self, factor_name: str) -> pd.Series | None:
        """
        Load factor time-series values from parquet file.

@@ -273,32 +269,32 @@ class StrategyOrchestrator:

        try:
            df = pd.read_parquet(str(parquet_path))
-            
+
            # Handle empty DataFrame
            if df.empty or len(df.columns) == 0:
                logger.warning(f"Empty parquet file: {parquet_path}")
                return None
-            
+
            # Handle MultiIndex (datetime, instrument)
            if isinstance(df.index, pd.MultiIndex):
                # Get the factor column name (should be the only column)
                factor_col = df.columns[0]
                # Extract EURUSD series
                try:
-                    series = df.xs('EURUSD', level='instrument')[factor_col]
+                    series = df.xs("EURUSD", level="instrument")[factor_col]
                except KeyError:
                    # Try alternative extraction
                    df_reset = df.reset_index()
-                    if 'instrument' in df_reset.columns:
-                        df_eur = df_reset[df_reset['instrument'] == 'EURUSD'].set_index('datetime')
+                    if "instrument" in df_reset.columns:
+                        df_eur = df_reset[df_reset["instrument"] == "EURUSD"].set_index("datetime")
                        series = df_eur[factor_col] if factor_col in df_eur.columns else df_eur.iloc[:, -1]
                    else:
                        series = df.iloc[:, 0]
            else:
                series = df.iloc[:, 0]
-            
+
            # Ensure numeric
-            series = pd.to_numeric(series, errors='coerce')
+            series = pd.to_numeric(series, errors="coerce")
            series.name = factor_name
            return series
        except Exception as e:
@@ -307,10 +303,10 @@ class StrategyOrchestrator:

    def generate_strategy_code(
        self,
-        factors: List[Dict[str, Any]],
+        factors: list[dict[str, Any]],
        strategy_name: str,
        max_retries: int = 3,
-    ) -> Optional[str]:
+    ) -> str | None:
        """
        Generate strategy code using LLM from factor combinations.

@@ -341,6 +337,12 @@ class StrategyOrchestrator:
            user_prompt = user_prompt.replace("{{ additional_context }}", f"Strategy name: {strategy_name}")
            user_prompt = user_prompt.replace("{{ trading_style }}", self.trading_style)
            user_prompt = user_prompt.replace("{{ min_sharpe }}", str(self.min_sharpe))
+
+            if "{{" in user_prompt:
+                unreplaced = [w for w in user_prompt.split() if "{{" in w]
+                logger.warning(
+                    f"Unreplaced template variables in prompt for '{strategy_name}': {unreplaced}"
+                )
            user_prompt = user_prompt.replace("{{ max_drawdown }}", str(self.max_drawdown))
            system_prompt = self.strategy_prompt.get("system", "")
        else:
@@ -371,12 +373,12 @@ class StrategyOrchestrator:
                    last_error = f"Attempt {attempt}: LLM returned empty or invalid code"
                    logger.warning(f"LLM attempt {attempt}/{max_retries} failed: {last_error}")
                except Exception as e:
-                    last_error = f"Attempt {attempt}: {str(e)}"
+                    last_error = f"Attempt {attempt}: {e!s}"
                    logger.warning(f"LLM attempt {attempt}/{max_retries} failed with exception: {e}")

            logger.warning(
                f"LLM strategy generation failed after {max_retries} attempts. "
-                f"Last error: {last_error}"
+                f"Last error: {last_error}",
            )

        # Fallback: generate template code programmatically
@@ -385,10 +387,10 @@ class StrategyOrchestrator:

    def _generate_with_llm(
        self,
-        context: Dict[str, Any],
+        context: dict[str, Any],
        attempt: int = 1,
-        feedback: Optional[str] = None,
-    ) -> Optional[str]:
+        feedback: str | None = None,
+    ) -> str | None:
        """
        Generate strategy code using LLM with APIBackend (same as Factor Coder).

@@ -406,7 +408,6 @@ class StrategyOrchestrator:
        str or None
            Validated Python strategy code, or None if invalid
        """
-        import json as json_module

        # Build user message with optional feedback
        user_content = context.get("user_prompt", "")
@@ -446,14 +447,13 @@ class StrategyOrchestrator:
                if self._validate_python_code(code):
                    logger.info(f"[DEBUG] Valid Python code extracted ({len(code)} chars)")
                    return code
-                else:
-                    logger.warning(f"JSON 'code' field contains invalid Python (attempt {attempt}). Preview: {code[:200]}")
+                logger.warning(f"JSON 'code' field contains invalid Python (attempt {attempt}). Preview: {code[:200]}")
            else:
                logger.warning(f"JSON parsed but no valid 'code' field found (attempt {attempt}). Keys: {list(json_data.keys())}")

        # === STEP 2: Fallback - Extract Python code block directly (like Factor Coder) ===
        import re
-        code_block_match = re.search(r'```python\s*\n(.*?)\n```', content, re.DOTALL)
+        code_block_match = re.search(r"```python\s*\n(.*?)\n```", content, re.DOTALL)
        if code_block_match:
            code = code_block_match.group(1).strip()
            if code and self._validate_python_code(code):
@@ -463,7 +463,7 @@ class StrategyOrchestrator:
        logger.warning(f"All extraction methods failed (attempt {attempt}). Response preview: {response[:200]}")
        return None

-    def _extract_json(self, content: str) -> Optional[Dict[str, Any]]:
+    def _extract_json(self, content: str) -> dict[str, Any] | None:
        """
        Extract JSON object from LLM response content.

@@ -491,7 +491,7 @@ class StrategyOrchestrator:
                pass

        # Strategy 2: Find ```json ... ``` blocks
-        json_block_match = re.search(r'```json\s*\n(.*?)\n```', content, re.DOTALL)
+        json_block_match = re.search(r"```json\s*\n(.*?)\n```", content, re.DOTALL)
        if json_block_match:
            try:
                return json_module.loads(json_block_match.group(1))
@@ -499,7 +499,7 @@ class StrategyOrchestrator:
                pass

        # Strategy 3: Find ```python ... ``` blocks (Qwen often puts JSON in python blocks)
-        python_block_match = re.search(r'```python\s*\n(.*?)\n```', content, re.DOTALL)
+        python_block_match = re.search(r"```python\s*\n(.*?)\n```", content, re.DOTALL)
        if python_block_match:
            block = python_block_match.group(1).strip()
            if block.startswith("{") and block.endswith("}"):
@@ -519,13 +519,13 @@ class StrategyOrchestrator:
                # Try to fix common JSON issues (trailing commas, unescaped newlines)
                try:
                    # Remove trailing commas before } or ]
-                    json_str_fixed = re.sub(r',\s*([}\]])', r'\1', json_str)
+                    json_str_fixed = re.sub(r",\s*([}\]])", r"\1", json_str)
                    return json_module.loads(json_str_fixed)
                except json_module.JSONDecodeError:
                    pass

        # Strategy 5: Find ``` ... ``` blocks (any language tag)
-        code_block_match = re.search(r'```\w*\s*\n(.*?)\n```', content, re.DOTALL)
+        code_block_match = re.search(r"```\w*\s*\n(.*?)\n```", content, re.DOTALL)
        if code_block_match:
            block = code_block_match.group(1).strip()
            if block.startswith("{") and block.endswith("}"):
@@ -536,7 +536,7 @@ class StrategyOrchestrator:

        return None

-    def _extract_code_from_json(self, json_data: Dict[str, Any]) -> Optional[str]:
+    def _extract_code_from_json(self, json_data: dict[str, Any]) -> str | None:
        """
        Extract Python code from parsed JSON data.

@@ -559,7 +559,7 @@ class StrategyOrchestrator:

        return None

-    def _extract_code_from_raw(self, content: str) -> Optional[str]:
+    def _extract_code_from_raw(self, content: str) -> str | None:
        """
        Extract Python code from raw (non-JSON) LLM response.

@@ -579,12 +579,12 @@ class StrategyOrchestrator:
        code = content.strip()

        # Try to find ```python blocks
-        python_match = re.search(r'```python\s*\n(.*?)\n```', code, re.DOTALL)
+        python_match = re.search(r"```python\s*\n(.*?)\n```", code, re.DOTALL)
        if python_match:
            code = python_match.group(1)
        else:
            # Try generic ``` blocks
-            block_match = re.search(r'```\s*\n(.*?)\n```', code, re.DOTALL)
+            block_match = re.search(r"```\s*\n(.*?)\n```", code, re.DOTALL)
            if block_match:
                code = block_match.group(1)

@@ -636,7 +636,7 @@ class StrategyOrchestrator:
        # Remove non-ASCII characters (emojis, etc.)
        code = code.encode("ascii", "ignore").decode("ascii").strip()

-        return code if code else None
+        return code or None

    def _validate_python_code(self, code: str) -> bool:
        """
@@ -663,14 +663,14 @@ class StrategyOrchestrator:
            logger.debug(f"Python syntax error: {e}")
            return False

-    def _generate_fallback_code(self, context: Dict[str, Any]) -> str:
+    def _generate_fallback_code(self, context: dict[str, Any]) -> str:
        """Generate fallback strategy code programmatically."""
        factor_names = context["factor_names"]
        style_config = "daytrading" if context["trading_style"] == "daytrading" else "swing"

        # Build factor assignment code
        factor_assignments = "\n    ".join(
-            [f'"{name}": factors["{name}"]' for name in factor_names if name != "timestamp"]
+            [f'"{name}": factors["{name}"]' for name in factor_names if name != "timestamp"],
        )

        code = f'''"""
@@ -717,8 +717,8 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
        return code

    def evaluate_strategy(
-        self, strategy_code: str, strategy_name: str, factors: List[Dict[str, Any]]
-    ) -> Dict[str, Any]:
+        self, strategy_code: str, strategy_name: str, factors: list[dict[str, Any]],
+    ) -> dict[str, Any]:
        """
        Evaluate a strategy by executing its code and calculating metrics.

@@ -755,23 +755,20 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                }

            # Align factor values with common index
-            if not factor_values:
-                df_factors = pd.DataFrame()
-            else:
-                # Find common index across all series
-                common_idx = None
-                for name, s in factor_values.items():
-                    if common_idx is None:
-                        common_idx = s.index
-                    else:
-                        common_idx = common_idx.intersection(s.index)
-                
-                if common_idx is not None and len(common_idx) > 100:
-                    df_factors = pd.DataFrame({
-                        name: s.reindex(common_idx) for name, s in factor_values.items()
-                    }).dropna()
+            # Find common index across all series
+            common_idx = None
+            for name, s in factor_values.items():
+                if common_idx is None:
+                    common_idx = s.index
                else:
-                    df_factors = pd.DataFrame()
+                    common_idx = common_idx.intersection(s.index)
+
+            if common_idx is not None and len(common_idx) > 100:
+                df_factors = pd.DataFrame({
+                    name: s.reindex(common_idx) for name, s in factor_values.items()
+                }).dropna()
+            else:
+                df_factors = pd.DataFrame()

            if len(df_factors) < 100:
                return {
@@ -783,8 +780,8 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)

            # Convert all factor columns to numeric
            for col in df_factors.columns:
-                df_factors[col] = pd.to_numeric(df_factors[col], errors='coerce')
-            
+                df_factors[col] = pd.to_numeric(df_factors[col], errors="coerce")
+
            # Forward-fill daily factors to match OHLCV 1-min index
            # Many factors are daily (1 value per day), need to ffill to 1-min
            # FIX 6: Track ffill ratio for data quality monitoring
@@ -799,11 +796,11 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                    f"[DEBUG] {strategy_name}: data quality: "
                    f"original_rows={original_len}, "
                    f"ffill_rows={len(df_factors) - original_len}, "
-                    f"ffill_ratio={ffill_ratio:.2%}"
+                    f"ffill_ratio={ffill_ratio:.2%}",
                )

            df_factors = df_factors.dropna()
-            
+
            if len(df_factors) < 1000:
                return {
                    "strategy_name": strategy_name,
@@ -811,54 +808,58 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                    "reason": f"Insufficient numeric data after conversion ({len(df_factors)} rows)",
                    "factors_used": factor_names,
                }
-            
+
            # close is already loaded above for ffill, reuse it
            # Reindex close to match factor index
            if close is not None:
                close = close.reindex(df_factors.index)
-            
+
            # Execute strategy code with factor data and close prices
            local_vars = {"factors": df_factors}
            if close is not None:
                local_vars["close"] = close
-            
+
            try:
                exec(strategy_code, {"np": np, "pd": pd, "numpy": np}, local_vars)
            except Exception as e:
+                import traceback
+                logger.error(
+                    f"Strategy code execution failed for '{strategy_name}': {e}\n"
+                    f"{traceback.format_exc()[-2000:]}"
+                )
                return {
                    "strategy_name": strategy_name,
                    "status": "rejected",
-                    "reason": f"Code execution error: {str(e)}",
+                    "reason": f"Code execution error: {e!s}",
                    "factors_used": factor_names,
                }

-            if "signal" not in local_vars:
+            signal = local_vars.get("signal")
+            if signal is None or (isinstance(signal, pd.Series) and signal.empty):
                return {
                    "strategy_name": strategy_name,
                    "status": "rejected",
-                    "reason": "Strategy did not produce 'signal' variable",
+                    "reason": "Strategy did not produce valid 'signal' variable",
                    "factors_used": factor_names,
                }

-            signal = local_vars["signal"]
-
            logger.info(
                f"[DEBUG] {strategy_name}: signal stats: "
                f"len={len(signal)}, "
                f"long={int((signal > 0).sum())}, "
                f"short={int((signal < 0).sum())}, "
                f"flat={int((signal == 0).sum())}, "
-                f"unique={signal.nunique()}"
+                f"unique={signal.nunique()}",
            )

            # Delegate all metric computation to the single source of truth.
            # Same formulas as every other backtest path in the repo.
            from rdagent.components.backtesting.vbt_backtest import (
-                backtest_signal,
                DEFAULT_TXN_COST_BPS,
+                backtest_signal_ftmo,
            )

-            close = self.load_ohlcv_close()
+            # Reuse the already-loaded close from above; create a synthetic proxy if unavailable
            if close is None:
                logger.warning("OHLCV data unavailable, using factor-mean proxy")
                proxy = df_factors.mean(axis=1).astype(float)
@@ -868,11 +869,10 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                close_for_bt = close.reindex(signal.index).ffill()

            txn_cost_bps = float(os.getenv("TXN_COST_BPS", DEFAULT_TXN_COST_BPS))
-            bt = backtest_signal(
+            bt = backtest_signal_ftmo(
                close=close_for_bt,
                signal=signal,
                txn_cost_bps=txn_cost_bps,
-                freq="1min",
            )

            if bt.get("status") != "success":
@@ -891,7 +891,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
            logger.info(
                f"[DEBUG] {strategy_name}: bt stats: "
                f"sharpe={sharpe:.4f} dd={max_dd:.4f} wr={win_rate:.4f} "
-                f"trades={num_real_trades} total_ret={bt['total_return']:.4%}"
+                f"trades={num_real_trades} total_ret={bt['total_return']:.4%}",
            )

            metrics = {
@@ -921,7 +921,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                    f"[DEBUG] {strategy_name}: rejection breakdown: "
                    f"sharpe={sharpe:.4f} (need>={self.min_sharpe}), "
                    f"dd={max_dd:.4f} (need>={self.max_drawdown}), "
-                    f"wr={win_rate:.4f} (need>={self.min_win_rate})"
+                    f"wr={win_rate:.4f} (need>={self.min_win_rate})",
                )
                metrics["reason"] = self._get_rejection_reason(sharpe, max_dd, win_rate)

@@ -933,7 +933,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
            return {
                "strategy_name": strategy_name,
                "status": "rejected",
-                "reason": f"Evaluation error: {str(e)}",
+                "reason": f"Evaluation error: {e!s}",
                "factors_used": [],
            }

@@ -952,7 +952,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
            reasons.append(f"Win Rate {win_rate:.2%} < {self.min_win_rate:.2%}")
        return "; ".join(reasons) if reasons else "Unknown"

-    def _generate_strategy_name(self, factors: List[Dict[str, Any]], idx: int) -> str:
+    def _generate_strategy_name(self, factors: list[dict[str, Any]], idx: int) -> str:
        """Generate a strategy name from its factors."""
        # Extract key words from factor names
        words = []
@@ -963,7 +963,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
            for p in parts:
                # Extract capitalized words
                cap_words = [w for w in p.split() if w[0:1].isupper()]
-                words.extend(cap_words if cap_words else [p])
+                words.extend(cap_words or [p])

        # Take up to 3 unique words
        unique_words = list(dict.fromkeys(words))[:3]
@@ -976,7 +976,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
        count: int = 10,
        workers: int = 2,  # Reduced from 4 to 2 to avoid LLM server overload
        progress_callback=None,
-    ) -> List[Dict[str, Any]]:
+    ) -> list[dict[str, Any]]:
        """
        Generate and evaluate trading strategies.

@@ -1029,7 +1029,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                        logger.info(
                            f"Strategy ACCEPTED: {result['strategy_name']} | "
                            f"Sharpe={result['sharpe_ratio']:.2f} | "
-                            f"DD={result['max_drawdown']:.2%}"
+                            f"DD={result['max_drawdown']:.2%}",
                        )
                    else:
                        # Also save rejected strategies for debugging
@@ -1037,7 +1037,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                        logger.warning(
                            f"Strategy REJECTED: {result['strategy_name']} - {result.get('reason', 'unknown')} | "
                            f"Sharpe={result.get('sharpe_ratio', 'N/A')} | "
-                            f"DD={result.get('max_drawdown', 'N/A')}"
+                            f"DD={result.get('max_drawdown', 'N/A')}",
                        )

                    if progress_callback:
@@ -1053,12 +1053,12 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)

        logger.info(
            f"Strategy generation complete: {strategies_accepted}/{strategies_generated} accepted "
-            f"({strategies_accepted/max(strategies_generated,1)*100:.1f}%)"
+            f"({strategies_accepted/max(strategies_generated,1)*100:.1f}%)",
        )

        return results

-    def _generate_strategy_configs(self, factors: List[Dict], count: int) -> List[List[Dict]]:
+    def _generate_strategy_configs(self, factors: list[dict], count: int) -> list[list[dict]]:
        """
        Generate strategy configurations from factor combinations.

@@ -1086,7 +1086,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
        np.random.shuffle(configs)
        return configs[: count * 2]  # Generate extras

-    def _generate_and_evaluate_single(self, idx: int, factors: List[Dict]) -> Dict[str, Any]:
+    def _generate_and_evaluate_single(self, idx: int, factors: list[dict]) -> dict[str, Any]:
        """Generate and evaluate a single strategy."""
        strategy_name = self._generate_strategy_name(factors, idx + 1)

@@ -1109,7 +1109,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
        # by finding optimal entry/exit thresholds, signal smoothing, etc.
        if self.use_optuna:
            initial_status = result.get("status", "rejected")
-            initial_sharpe = result.get("sharpe_ratio", float('-inf'))
+            initial_sharpe = result.get("sharpe_ratio", float("-inf"))
            logger.info(f"Running Optuna optimization for {strategy_name} (initial: {initial_status}, Sharpe={initial_sharpe:.4f})...")
            optimizer = OptunaOptimizer(n_trials=self.optuna_trials)

@@ -1118,7 +1118,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)

            if factor_values is not None:
                optimized = optimizer.optimize_strategy(result, factor_values)
-                optimized_sharpe = optimized.get("sharpe_ratio", float('-inf'))
+                optimized_sharpe = optimized.get("sharpe_ratio", float("-inf"))
                optimized_status = optimized.get("status", "rejected")
                best_params = optimized.get("best_params", {})

@@ -1127,14 +1127,14 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                    improvement = optimized_sharpe - initial_sharpe
                    logger.info(
                        f"Optuna {'RESCUED' if optimized_status == 'accepted' and initial_status == 'rejected' else 'improved'} "
-                        f"{strategy_name}: Sharpe {initial_sharpe:.4f} → {optimized_sharpe:.4f} (+{improvement:.4f})"
+                        f"{strategy_name}: Sharpe {initial_sharpe:.4f} → {optimized_sharpe:.4f} (+{improvement:.4f})",
                    )

                    # Re-evaluate with best parameters to get comparable metrics
                    if best_params:
                        patched_code = self._patch_strategy_code(code, best_params)
                        re_eval = self._evaluate_with_patched_code(patched_code, strategy_name, factors)
-                        if re_eval.get("sharpe_ratio", float('-inf')) > initial_sharpe:
+                        if re_eval.get("sharpe_ratio", float("-inf")) > initial_sharpe:
                            result.update(re_eval)
                            result["code"] = patched_code
                            result["best_params"] = best_params
@@ -1143,7 +1143,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                                result.pop("reason", None)
                            logger.info(
                                f"Re-evaluated {strategy_name} with best params: "
-                                f"Sharpe {initial_sharpe:.4f} → {re_eval.get('sharpe_ratio', 0):.4f}"
+                                f"Sharpe {initial_sharpe:.4f} → {re_eval.get('sharpe_ratio', 0):.4f}",
                            )
                        else:
                            result.update(optimized)
@@ -1163,7 +1163,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)

        return result

-    def _prepare_factor_values(self, factors: List[Dict]) -> Optional[pd.DataFrame]:
+    def _prepare_factor_values(self, factors: list[dict]) -> pd.DataFrame | None:
        """Prepare factor values DataFrame for Optuna optimization."""
        factor_values = {}
        for f in factors:
@@ -1182,7 +1182,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
            return df.dropna()
        return None

-    def _patch_strategy_code(self, code: str, params: Dict[str, Any]) -> str:
+    def _patch_strategy_code(self, code: str, params: dict[str, Any]) -> str:
        """Patch strategy code with Optuna's best parameters."""
        import re
        patched = code
@@ -1193,26 +1193,26 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
        signal_window = params.get("signal_window", 3)

        param_patterns = [
-            (r'entry_thresh\s*=\s*[\d.]+', f'entry_thresh = {entry_thresh}'),
-            (r'exit_thresh\s*=\s*[\d.]+', f'exit_thresh = {exit_thresh}'),
-            (r'window\s*=\s*\d+', f'window = {zscore_window}'),
-            (r'signal_window\s*=\s*\d+', f'signal_window = {signal_window}'),
+            (r"entry_thresh\s*=\s*[\d.]+", f"entry_thresh = {entry_thresh}"),
+            (r"exit_thresh\s*=\s*[\d.]+", f"exit_thresh = {exit_thresh}"),
+            (r"window\s*=\s*\d+", f"window = {zscore_window}"),
+            (r"signal_window\s*=\s*\d+", f"signal_window = {signal_window}"),
        ]
        for pattern, replacement in param_patterns:
            patched = re.sub(pattern, replacement, patched)

        # Patch .rolling(N) calls for common window sizes
-        rolling_pattern = r'\.rolling\((\d+)\)'
+        rolling_pattern = r"\.rolling\((\d+)\)"
        def replace_rolling(match):
            val = int(match.group(1))
            if val in (20, 30, 50, 100, 200):
-                return f'.rolling({zscore_window})'
+                return f".rolling({zscore_window})"
            return match.group(0)
        patched = re.sub(rolling_pattern, replace_rolling, patched)

        return patched

-    def _evaluate_with_patched_code(self, patched_code: str, strategy_name: str, factors: List[Dict]) -> Dict[str, Any]:
+    def _evaluate_with_patched_code(self, patched_code: str, strategy_name: str, factors: list[dict]) -> dict[str, Any]:
        """Re-evaluate strategy with patched parameters using full OHLCV backtest."""
        try:
            factor_names = [f["factor_name"] for f in factors if f["factor_name"] != "timestamp"]
@@ -1223,7 +1223,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                    factor_values[fname] = series

            if not factor_values:
-                return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+                return {"sharpe_ratio": float("-inf"), "status": "rejected"}

            common_idx = None
            for name, s in factor_values.items():
@@ -1233,14 +1233,14 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
                    common_idx = common_idx.intersection(s.index)

            if common_idx is None or len(common_idx) < 100:
-                return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+                return {"sharpe_ratio": float("-inf"), "status": "rejected"}

            df_factors = pd.DataFrame({
                name: s.reindex(common_idx) for name, s in factor_values.items()
            }).dropna()

            for col in df_factors.columns:
-                df_factors[col] = pd.to_numeric(df_factors[col], errors='coerce')
+                df_factors[col] = pd.to_numeric(df_factors[col], errors="coerce")

            close = self.load_ohlcv_close()
            if close is not None:
@@ -1248,7 +1248,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)

            df_factors = df_factors.dropna()
            if len(df_factors) < 1000:
-                return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+                return {"sharpe_ratio": float("-inf"), "status": "rejected"}

            if close is not None:
                close = close.reindex(df_factors.index)
@@ -1257,30 +1257,29 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)
            try:
                exec(patched_code, {"np": np, "pd": pd, "numpy": np}, local_vars)
            except Exception:
-                return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+                return {"sharpe_ratio": float("-inf"), "status": "rejected"}

            if "signal" not in local_vars:
-                return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+                return {"sharpe_ratio": float("-inf"), "status": "rejected"}

            signal = local_vars["signal"]

            from rdagent.components.backtesting.vbt_backtest import (
-                backtest_signal,
                DEFAULT_TXN_COST_BPS,
+                backtest_signal_ftmo,
            )

            close_for_bt = close.reindex(signal.index).ffill() if close is not None else None
            if close_for_bt is None:
-                return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+                return {"sharpe_ratio": float("-inf"), "status": "rejected"}

-            bt = backtest_signal(
+            bt = backtest_signal_ftmo(
                close=close_for_bt,
                signal=signal,
                txn_cost_bps=float(os.getenv("TXN_COST_BPS", DEFAULT_TXN_COST_BPS)),
-                freq="1min",
            )
            if bt.get("status") != "success":
-                return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+                return {"sharpe_ratio": float("-inf"), "status": "rejected"}

            sharpe = bt["sharpe"]
            max_dd = bt["max_drawdown"]
@@ -1309,9 +1308,9 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)

        except Exception as e:
            logger.debug(f"Re-evaluation failed for {strategy_name}: {e}")
-            return {"sharpe_ratio": float('-inf'), "status": "rejected"}
+            return {"sharpe_ratio": float("-inf"), "status": "rejected"}

-    def _save_strategy(self, result: Dict[str, Any]) -> None:
+    def _save_strategy(self, result: dict[str, Any]) -> None:
        """Save accepted strategy to JSON file."""
        timestamp = int(time.time())
        safe_name = result["strategy_name"].replace("/", "_").replace(" ", "_")[:60]
@@ -1327,7 +1326,7 @@ signal = signal.rolling(window=3, min_periods=1).mean().round().astype(int)

        logger.info(f"Saved strategy to {filepath}")

-    def get_strategy_summary(self, results: List[Dict[str, Any]]) -> Dict[str, Any]:
+    def get_strategy_summary(self, results: list[dict[str, Any]]) -> dict[str, Any]:
        """
        Generate summary statistics from strategy generation results.

@@ -85,13 +85,16 @@ def load_and_process_one_pdf_by_azure_document_intelligence(


 def load_and_process_pdfs_by_azure_document_intelligence(path: Path) -> dict[str, str]:
-    assert RD_AGENT_SETTINGS.azure_document_intelligence_key is not None
-    assert RD_AGENT_SETTINGS.azure_document_intelligence_endpoint is not None
+    if RD_AGENT_SETTINGS.azure_document_intelligence_key is None:
+        raise AssertionError("azure_document_intelligence_key must be set")
+    if RD_AGENT_SETTINGS.azure_document_intelligence_endpoint is None:
+        raise AssertionError("azure_document_intelligence_endpoint must be set")

    content_dict = {}
    ab_path = path.resolve()
    if ab_path.is_file():
-        assert ".pdf" in ab_path.suffixes, "The file must be a PDF file."
+        if ".pdf" not in ab_path.suffixes:
+            raise ValueError("The file must be a PDF file.")
        proc = load_and_process_one_pdf_by_azure_document_intelligence
        content_dict[str(ab_path)] = proc(
            ab_path,
@@ -24,7 +24,8 @@ class UndirectedNode(Node):
        super().__init__(content, label, embedding)
        self.neighbors: set[UndirectedNode] = set()
        self.appendix = appendix  # appendix stores any additional information
-        assert isinstance(content, str), "content must be a string"
+        if not isinstance(content, str):
+            raise TypeError("content must be a string")

    def add_neighbor(self, node: UndirectedNode) -> None:
        self.neighbors.add(node)
@@ -96,7 +97,8 @@ class Graph(KnowledgeBase):
                APIBackend().create_embedding(input_content=contents[i : i + size]),
            )

-        assert len(nodes) == len(embeddings), "nodes' length must equals embeddings' length"
+        if len(nodes) != len(embeddings):
+            raise ValueError("nodes' length must equal embeddings' length")
        for node, embedding in zip(nodes, embeddings):
            node.embedding = embedding
        return nodes
@@ -252,7 +254,8 @@ class UndirectedGraph(Graph):

        """
        min_nodes_count = 2
-        assert len(nodes) >= min_nodes_count, "nodes length must >=2"
+        if len(nodes) < min_nodes_count:
+            raise ValueError("nodes length must >=2")
        intersection = None

        for node in nodes:
@@ -87,7 +87,8 @@ class ModelWsLoader(WsLoader[ModelTask, ModelFBWorkspace]):
        self.path = Path(path)

    def load(self, task: ModelTask) -> ModelFBWorkspace:
-        assert task.name is not None
+        if task.name is None:
+            raise AssertionError("task.name should not be None")
        mti = ModelFBWorkspace(task)
        mti.prepare()
        with open(self.path / f"{task.name}.py", "r") as f:
@@ -4,6 +4,7 @@ import functools
 import importlib
 import json
 import multiprocessing as mp
+import os
 import pickle
 import random
 from collections.abc import Callable
@@ -82,10 +83,24 @@ def import_class(class_path: str) -> Any:
    Returns
    -------
        class of `class_path`
+
+    Raises
+    ------
+    ImportError
+        If module or class cannot be found.
    """
-    module_path, class_name = class_path.rsplit(".", 1)
-    module = importlib.import_module(module_path)
-    return getattr(module, class_name)
+    try:
+        module_path, class_name = class_path.rsplit(".", 1)
+    except ValueError:
+        raise ImportError(f"Invalid class path: {class_path!r}")
+    try:
+        module = importlib.import_module(module_path)
+    except ModuleNotFoundError as e:
+        raise ImportError(f"Module not found: {module_path!r}") from e
+    try:
+        return getattr(module, class_name)
+    except AttributeError as e:
+        raise ImportError(f"Class not found: {class_name!r} in {module_path!r}") from e


 class CacheSeedGen:
@@ -208,3 +223,13 @@ def cache_with_pickle(hash_func: Callable, post_process_func: Callable | None =
        return cache_wrapper

    return cache_decorator
+
+
+def safe_resolve_path(user_path: Path, safe_root: Path | None = None) -> Path:
+    if safe_root is not None:
+        root_real = os.path.realpath(str(safe_root.expanduser()))
+        path_real = os.path.realpath(str(user_path.expanduser()))  # nosec B614 — validated against safe_root below
+        if not (path_real == root_real or path_real.startswith(root_real + os.sep)):
+            raise ValueError(f"Path {user_path} resolves to {path_real}, outside allowed root {safe_root}")
+        return Path(path_real)
+    return user_path.expanduser().resolve()
@@ -27,6 +27,7 @@ Usage:
 from __future__ import annotations

 import json as _json
+import logging
 import sys
 import threading
 from contextlib import contextmanager
@@ -36,21 +37,24 @@ from typing import Any

 from loguru import logger as _root

-# ── paths ─────────────────────────────────────────────────────────────────────
+# ── paths ─────────────────────────────────────────────────────────────────────────────────
 LOGS_ROOT: Path = Path(__file__).parent.parent.parent / "logs"

-# ── format ────────────────────────────────────────────────────────────────────
+# ── format ────────────────────────────────────────────────────────────────────────────────
 _FILE_FMT = (
    "{time:YYYY-MM-DD HH:mm:ss.SSS} | {level: <8} | {extra[cmd]: <18} | {message}"
 )

-# ── internal state ─────────────────────────────────────────────────────────────
+# ── internal state ─────────────────────────────────────────────────────────────────────────────
 _registered: set[str] = set()   # command keys that already have a file sink
 _all_added: bool = False         # whether the combined all.log sink is active
 _llm_log_lock = threading.Lock()  # guards concurrent writes to llm_calls.jsonl

+# Maximum characters stored per field in llm_calls.jsonl to prevent GB-scale files.
+_LLM_CALL_MAX_CHARS = 500

-# ── helpers ───────────────────────────────────────────────────────────────────
+
+# ── helpers ────────────────────────────────────────────────────────────────────────────────

 def _today_dir() -> Path:
    d = LOGS_ROOT / datetime.now().strftime("%Y-%m-%d")
@@ -79,7 +83,7 @@ def _banner(log, title: str, meta: dict[str, Any]) -> None:
    log.info(sep)


-# ── public API ────────────────────────────────────────────────────────────────
+# ── public API ──────────────────────────────────────────────────────────────────────────────

 def log_llm_call(
    system: str | None,
@@ -88,16 +92,19 @@ def log_llm_call(
    start_time: Any = None,
    end_time: Any = None,
 ) -> None:
-    """Append one complete LLM call to logs/YYYY-MM-DD/llm_calls.jsonl.
+    """Append one LLM call summary to logs/YYYY-MM-DD/llm_calls.jsonl.
+
+    Prompt/response content is capped at _LLM_CALL_MAX_CHARS to prevent
+    GB-scale log files from long-running loops.

    Each line is a self-contained JSON object so the file is grep/jq-friendly:
      jq 'select(.duration_ms > 5000)' logs/2026-04-17/llm_calls.jsonl
    """
    entry: dict[str, Any] = {
        "ts": datetime.now().isoformat(timespec="milliseconds"),
-        "system": system or "",
-        "user": user,
-        "response": response,
+        "system": (system or "")[:_LLM_CALL_MAX_CHARS],
+        "user": user[:_LLM_CALL_MAX_CHARS],
+        "response": response[:_LLM_CALL_MAX_CHARS],
    }
    if start_time is not None and end_time is not None:
        try:
@@ -130,13 +137,13 @@ def setup(command: str, **context: Any):
    key = command.lower()

    if key not in _registered:
-        # Per-command rotating file
        _root.add(
            str(log_dir / f"{key}.log"),
            format=_FILE_FMT,
            filter=lambda r, k=key: r["extra"].get("cmd", "").lower() == k,
-            rotation="00:00",      # new file at midnight
-            retention="30 days",
+            rotation="50 MB",
+            compression="gz",
+            retention="7 days",
            encoding="utf-8",
            enqueue=True,
            backtrace=False,
@@ -145,13 +152,13 @@ def setup(command: str, **context: Any):
        _registered.add(key)

    if not _all_added:
-        # Combined log — all commands
        _root.add(
            str(log_dir / "all.log"),
            format=_FILE_FMT,
            filter=lambda r: "cmd" in r["extra"],
-            rotation="00:00",
-            retention="60 days",
+            rotation="100 MB",
+            compression="gz",
+            retention="7 days",
            encoding="utf-8",
            enqueue=True,
            backtrace=False,
@@ -0,0 +1,306 @@
+import argparse
+import json
+import pickle  # nosec
+import re
+import time
+from pathlib import Path
+
+import streamlit as st
+from streamlit import session_state
+
+from rdagent.log.ui.conf import UI_SETTING
+from rdagent.log.utils import extract_evoid, extract_loopid_func_name
+
+st.set_page_config(layout="wide", page_title="debug_llm", page_icon="🎓", initial_sidebar_state="expanded")
+
+# 获取 log_path 参数
+parser = argparse.ArgumentParser(description="RD-Agent Streamlit App")
+parser.add_argument("--log_dir", type=str, help="Path to the log directory")
+args = parser.parse_args()
+
+
+def get_folders_sorted(log_path):
+    """缓存并返回排序后的文件夹列表，并加入进度打印"""
+    with st.spinner("正在加载文件夹列表..."):
+        folders = sorted(
+            (folder for folder in log_path.iterdir() if folder.is_dir() and list(folder.iterdir())),
+            key=lambda folder: folder.stat().st_mtime,
+            reverse=True,
+        )
+        st.write(f"找到 {len(folders)} 个文件夹")
+    return [folder.name for folder in folders]
+
+
+if UI_SETTING.enable_cache:
+    get_folders_sorted = st.cache_data(get_folders_sorted)
+
+
+# 设置主日志路径
+main_log_path = Path(args.log_dir) if args.log_dir else Path("./log")
+if not main_log_path.exists():
+    st.error(f"Log dir {main_log_path} does not exist!")
+    st.stop()
+
+if "data" not in session_state:
+    session_state.data = []
+if "log_path" not in session_state:
+    session_state.log_path = None
+
+tlist = []
+
+
+def load_data():
+    """加载数据到 session_state 并显示进度"""
+    log_file = main_log_path / session_state.log_path / "debug_llm.pkl"
+    try:
+        with st.spinner(f"正在加载数据文件 {log_file}..."):
+            start_time = time.time()
+            with open(log_file, "rb") as f:
+                session_state.data = pickle.load(f, encoding="utf-8")  # nosec
+            st.success(f"数据加载完成！耗时 {time.time() - start_time:.2f} 秒")
+            st.session_state["current_loop"] = 1
+    except Exception as e:
+        session_state.data = [{"error": str(e)}]
+        st.error(f"加载数据失败: {e}")
+
+
+# UI - Sidebar
+with st.sidebar:
+    st.markdown(":blue[**Log Path**]")
+    manually = st.toggle("Manual Input")
+    if manually:
+        st.text_input("log path", key="log_path", label_visibility="collapsed")
+    else:
+        folders = get_folders_sorted(main_log_path)
+        st.selectbox(f"**Select from {main_log_path.absolute()}**", folders, key="log_path")  # nosec B608 — not SQL, Bandit false positive on "Select" in UI label
+
+    if st.button("Refresh Data"):
+        load_data()
+        st.rerun()
+
+
+# Helper functions
+def show_text(text, lang=None):
+    """显示文本代码块"""
+    if lang:
+        st.code(text, language=lang, wrap_lines=True)
+    elif "\n" in text:
+        st.code(text, language="python", wrap_lines=True)
+    else:
+        st.code(text, language="html", wrap_lines=True)
+
+
+def highlight_prompts_uri(uri):
+    """高亮 URI 的格式"""
+    parts = uri.split(":")
+    return f"**{parts[0]}:**:green[**{parts[1]}**]"
+
+
+# Display Data
+progress_text = st.empty()
+progress_bar = st.progress(0)
+
+# 每页展示一个 Loop
+LOOPS_PER_PAGE = 1
+
+# 获取所有的 Loop ID
+loop_groups = {}
+for i, d in enumerate(session_state.data):
+    tag = d["tag"]
+    loop_id, _ = extract_loopid_func_name(tag)
+    if loop_id:
+        if loop_id not in loop_groups:
+            loop_groups[loop_id] = []
+        loop_groups[loop_id].append(d)
+
+# 按 Loop ID 排序
+sorted_loop_ids = sorted(loop_groups.keys(), key=int)  # 假设 Loop ID 是数字
+total_loops = len(sorted_loop_ids)
+total_pages = total_loops  # 每页展示一个 Loop
+
+
+# simple display
+# FIXME: Delete this simple UI if trace have tag(evo_id & loop_id)
+# with st.sidebar:
+#     start = int(st.text_input("start", 0))
+#     end = int(st.text_input("end", 100))
+# for m in session_state.data[start:end]:
+#     if "tpl" in m["tag"]:
+#         obj = m["obj"]
+#         uri = obj["uri"]
+#         tpl = obj["template"]
+#         cxt = obj["context"]
+#         rd = obj["rendered"]
+#         with st.expander(highlight_prompts_uri(uri), expanded=False, icon="⚙️"):
+#             t1, t2, t3 = st.tabs([":green[**Rendered**]", ":blue[**Template**]", ":orange[**Context**]"])
+#             with t1:
+#                 show_text(rd)
+#             with t2:
+#                 show_text(tpl, lang="django")
+#             with t3:
+#                 st.json(cxt)
+#     if "llm" in m["tag"]:
+#         obj = m["obj"]
+#         system = obj.get("system", None)
+#         user = obj["user"]
+#         resp = obj["resp"]
+#         with st.expander(f"**LLM**", expanded=False, icon="🤖"):
+#             t1, t2, t3 = st.tabs([":green[**Response**]", ":blue[**User**]", ":orange[**System**]"])
+#             with t1:
+#                 try:
+#                     rdict = json.loads(resp)
+#                     if "code" in rdict:
+#                         code = rdict["code"]
+#                         st.markdown(":red[**Code in response dict:**]")
+#                         st.code(code, language="python", wrap_lines=True, line_numbers=True)
+#                         rdict.pop("code")
+#                     elif "spec" in rdict:
+#                         spec = rdict["spec"]
+#                         st.markdown(":red[**Spec in response dict:**]")
+#                         st.markdown(spec)
+#                         rdict.pop("spec")
+#                     else:
+#                         # show model codes
+#                         showed_keys = []
+#                         for k, v in rdict.items():
+#                             if k.startswith("model_") and k.endswith(".py"):
+#                                 st.markdown(f":red[**{k}**]")
+#                                 st.code(v, language="python", wrap_lines=True, line_numbers=True)
+#                                 showed_keys.append(k)
+#                         for k in showed_keys:
+#                             rdict.pop(k)
+#                     st.write(":red[**Other parts (except for the code or spec) in response dict:**]")
+#                     st.json(rdict)
+#                 except:
+#                     st.json(resp)
+#             with t2:
+#                 show_text(user)
+#             with t3:
+#                 show_text(system or "No system prompt available")
+
+
+if total_pages:
+    # 初始化 current_loop
+    if "current_loop" not in st.session_state:
+        st.session_state["current_loop"] = 1
+
+    # Loop 导航按钮
+    col1, col2, col3, col4, col5 = st.sidebar.columns([1.2, 1, 2, 1, 1.2])
+
+    with col1:
+        if st.button("|<"):  # 首页
+            st.session_state["current_loop"] = 1
+    with col2:
+        if st.button("<") and st.session_state["current_loop"] > 1:  # 上一页
+            st.session_state["current_loop"] -= 1
+    with col3:
+        # 下拉列表显示所有 Loop
+        st.session_state["current_loop"] = st.selectbox(
+            "选择 Loop",
+            options=list(range(1, total_loops + 1)),
+            index=st.session_state["current_loop"] - 1,  # 默认选中当前 Loop
+            label_visibility="collapsed",  # 隐藏标签
+        )
+    with col4:
+        if st.button("\>") and st.session_state["current_loop"] < total_loops:  # 下一页
+            st.session_state["current_loop"] += 1
+    with col5:
+        if st.button("\>|"):  # 最后一页
+            st.session_state["current_loop"] = total_loops
+
+    # 获取当前 Loop
+    current_loop = st.session_state["current_loop"]
+
+    # 渲染当前 Loop 数据
+    loop_id = sorted_loop_ids[current_loop - 1]
+    progress_text = st.empty()
+    progress_text.text(f"正在处理 Loop {loop_id}...")
+    progress_bar.progress(current_loop / total_loops, text=f"Loop :green[**{current_loop}**] / {total_loops}")
+
+    # 渲染 Loop Header
+    loop_anchor = f"Loop_{loop_id}"
+    if loop_anchor not in tlist:
+        tlist.append(loop_anchor)
+        st.header(loop_anchor, anchor=loop_anchor, divider="blue")
+
+    # 渲染当前 Loop 的所有数据
+    loop_data = loop_groups[loop_id]
+    for d in loop_data:
+        tag = d["tag"]
+        obj = d["obj"]
+        _, func_name = extract_loopid_func_name(tag)
+        evo_id = extract_evoid(tag)
+
+        func_anchor = f"loop_{loop_id}.{func_name}"
+        if func_anchor not in tlist:
+            tlist.append(func_anchor)
+            st.header(f"in *{func_name}*", anchor=func_anchor, divider="green")
+
+        evo_anchor = f"loop_{loop_id}.evo_step_{evo_id}"
+        if evo_id and evo_anchor not in tlist:
+            tlist.append(evo_anchor)
+            st.subheader(f"evo_step_{evo_id}", anchor=evo_anchor, divider="orange")
+
+        # 根据 tag 渲染内容
+        if "debug_exp_gen" in tag:
+            with st.expander(
+                f"Exp in :violet[**{obj.experiment_workspace.workspace_path}**]", expanded=False, icon="🧩"
+            ):
+                st.write(obj)
+        elif "debug_tpl" in tag:
+            uri = obj["uri"]
+            tpl = obj["template"]
+            cxt = obj["context"]
+            rd = obj["rendered"]
+            with st.expander(highlight_prompts_uri(uri), expanded=False, icon="⚙️"):
+                t1, t2, t3 = st.tabs([":green[**Rendered**]", ":blue[**Template**]", ":orange[**Context**]"])
+                with t1:
+                    show_text(rd)
+                with t2:
+                    show_text(tpl, lang="django")
+                with t3:
+                    st.json(cxt)
+        elif "debug_llm" in tag:
+            system = obj.get("system", None)
+            user = obj["user"]
+            resp = obj["resp"]
+            with st.expander(f"**LLM**", expanded=False, icon="🤖"):
+                t1, t2, t3 = st.tabs([":green[**Response**]", ":blue[**User**]", ":orange[**System**]"])
+                with t1:
+                    try:
+                        rdict = json.loads(resp)
+                        if "code" in rdict:
+                            code = rdict["code"]
+                            st.markdown(":red[**Code in response dict:**]")
+                            st.code(code, language="python", wrap_lines=True, line_numbers=True)
+                            rdict.pop("code")
+                        elif "spec" in rdict:
+                            spec = rdict["spec"]
+                            st.markdown(":red[**Spec in response dict:**]")
+                            st.markdown(spec)
+                            rdict.pop("spec")
+                        else:
+                            # show model codes
+                            showed_keys = []
+                            for k, v in rdict.items():
+                                if k.startswith("model_") and k.endswith(".py"):
+                                    st.markdown(f":red[**{k}**]")
+                                    st.code(v, language="python", wrap_lines=True, line_numbers=True)
+                                    showed_keys.append(k)
+                            for k in showed_keys:
+                                rdict.pop(k)
+                        st.write(":red[**Other parts (except for the code or spec) in response dict:**]")
+                        st.json(rdict)
+                    except:
+                        st.json(resp)
+                with t2:
+                    show_text(user)
+                with t3:
+                    show_text(system or "No system prompt available")
+
+    progress_text.text("当前 Loop 数据处理完成！")
+
+    # Sidebar TOC
+    with st.sidebar:
+        toc = "\n".join([f"- [{t}](#{t})" if t.startswith("L") else f"  - [{t.split('.')[1]}](#{t})" for t in tlist])
+        st.markdown(toc, unsafe_allow_html=True)
@@ -541,7 +541,8 @@ class APIBackend(ABC):
        **kwargs,
    ) -> str | list[list[float]]:
        """This function to share operation between embedding and chat completion"""
-        assert not (chat_completion and embedding), "chat_completion and embedding cannot be True at the same time"
+        if chat_completion and embedding:
+            raise ValueError("chat_completion and embedding cannot be True at the same time")
        max_retry = LLM_SETTINGS.max_retry if LLM_SETTINGS.max_retry is not None else max_retry
        timeout_count = 0
        violation_count = 0
@@ -720,7 +721,13 @@ class APIBackend(ABC):

            if finish_reason is None or finish_reason != "length":
                break  # we get a full response now.
-            new_messages.append({"role": "assistant", "content": response})
+            # Merge into the previous assistant message if there already is one at the end.
+            # Appending a second consecutive assistant message causes llama-server to return 400
+            # ("Cannot have 2 or more assistant messages at the end of the list").
+            if new_messages and new_messages[-1]["role"] == "assistant":
+                new_messages[-1]["content"] += response
+            else:
+                new_messages.append({"role": "assistant", "content": response})
        else:
            raise RuntimeError(f"Failed to continue the conversation after {try_n} retries.")

@@ -36,16 +36,18 @@ def get_agent_model() -> OpenAIChatModel:

    """
    backend = APIBackend()
-    assert isinstance(backend, LiteLLMAPIBackend), "Only LiteLLMAPIBackend is supported"
+    if not isinstance(backend, LiteLLMAPIBackend):
+        raise TypeError("Only LiteLLMAPIBackend is supported")

    compl_kwargs = backend.get_complete_kwargs()

    selected_model = compl_kwargs["model"]

    _, custom_llm_provider, _, _ = get_llm_provider(selected_model)
-    assert (
-        custom_llm_provider in PROVIDER_TO_ENV_MAP
-    ), f"Provider {custom_llm_provider} not supported. Please add it into `PROVIDER_TO_ENV_MAP`"
+    if custom_llm_provider not in PROVIDER_TO_ENV_MAP:
+        raise ValueError(
+            f"Provider {custom_llm_provider} not supported. Please add it into `PROVIDER_TO_ENV_MAP`"
+        )
    prefix = PROVIDER_TO_ENV_MAP[custom_llm_provider]
    api_key = os.getenv(f"{prefix}_API_KEY", None)
    api_base = os.getenv(f"{prefix}_API_BASE", None)
@@ -268,7 +268,8 @@ class JsonReducer(DataReducer):
                parent[key] = sampled  # type: ignore  # parent 是 list，key 是 index, list.__setitem__(key, sampled)
            self.sampled_files.extend([self.extract_filename(i) for i in sampled])
            break
-        assert len(self.sampled_files) > 0
+        if len(self.sampled_files) <= 0:
+            raise AssertionError("sampled_files must contain at least one file")
        return data

    def _find_all_lists(
@@ -7,8 +7,10 @@ from sklearn.metrics import roc_auc_score
 def prepare_for_auroc_metric(submission: pd.DataFrame, answers: pd.DataFrame, id_col: str, target_col: str) -> dict:

    # Answers checks
-    assert id_col in answers.columns, f"answers dataframe should have an {id_col} column"
-    assert target_col in answers.columns, f"answers dataframe should have a {target_col} column"
+    if id_col not in answers.columns:
+        raise InvalidSubmissionError(f"answers dataframe should have an {id_col} column")
+    if target_col not in answers.columns:
+        raise InvalidSubmissionError(f"answers dataframe should have a {target_col} column")

    # Submission checks
    if id_col not in submission.columns:
@@ -1,7 +1,8 @@
 from pathlib import Path

 # Check if our submission file exists
-assert Path("submission.csv").exists(), "Error: submission.csv not found"
+if not Path("submission.csv").exists():
+    raise FileNotFoundError("Error: submission.csv not found")

 submission_lines = Path("submission.csv").read_text().splitlines()
 test_lines = Path("submission_test.csv").read_text().splitlines()
@@ -22,7 +22,8 @@ def prepare_for_metric(submission: pd.DataFrame, answers: pd.DataFrame) -> dict:
    if "price" not in submission.columns:
        raise InvalidSubmissionError("Submission DataFrame must contain 'price' columns.")

-    assert "price" in answers.columns, "Answers DataFrame must contain 'price' columns."
+    if "price" not in answers.columns:
+        raise InvalidSubmissionError("Answers DataFrame must contain 'price' columns.")

    if len(submission) != len(answers):
        raise InvalidSubmissionError("Submission must be the same length as the answers.")
@@ -1,7 +1,8 @@
 from pathlib import Path

 # Check if our submission file exists
-assert Path("submission.csv").exists(), "Error: submission.csv not found"
+if not Path("submission.csv").exists():
+    raise FileNotFoundError("Error: submission.csv not found")

 submission_lines = Path("submission.csv").read_text().splitlines()  # 自动生成的
 test_lines = Path("submission_test.csv").read_text().splitlines()  # test.csv
@@ -56,14 +56,17 @@ sparse.save_npz(public / "test" / "X.npz", X_test)
 sparse.save_npz(public / "train" / "X.npz", X_train)
 df_train.to_csv(public / "train" / "ARF_12h.csv", index=False)

-assert (
-    X_train.shape[0] == df_train.shape[0]
-), f"Mismatch: X_train rows ({X_train.shape[0]}) != df_train rows ({df_train.shape[0]})"
-assert (
-    X_test.shape[0] == df_test.shape[0]
-), f"Mismatch: X_test rows ({X_test.shape[0]}) != df_test rows ({df_test.shape[0]})"
-assert df_test.shape[1] == 2, "Public test set should have 2 columns"
-assert df_train.shape[1] == 3, "Public train set should have 3 columns"
-assert len(df_train) + len(df_test) == len(
-    df_label
-), "Length of new_train and new_test should equal length of old_train"
+if X_train.shape[0] != df_train.shape[0]:
+    raise ValueError(
+        f"Mismatch: X_train rows ({X_train.shape[0]}) != df_train rows ({df_train.shape[0]})"
+    )
+if X_test.shape[0] != df_test.shape[0]:
+    raise ValueError(
+        f"Mismatch: X_test rows ({X_test.shape[0]}) != df_test rows ({df_test.shape[0]})"
+    )
+if df_test.shape[1] != 2:
+    raise ValueError("Public test set should have 2 columns")
+if df_train.shape[1] != 3:
+    raise ValueError("Public train set should have 3 columns")
+if len(df_train) + len(df_test) != len(df_label):
+    raise ValueError("Length of new_train and new_test should equal length of old_train")
@@ -25,11 +25,12 @@ def prepare(raw: Path, public: Path, private: Path):
    new_test.to_csv(public / "test.csv", index=False)

    # Checks
-    assert new_test.shape[1] == 12, "Public test set should have 12 columns"
-    assert new_train.shape[1] == 13, "Public train set should have 13 columns"
-    assert len(new_train) + len(new_test) == len(
-        old_train
-    ), "Length of new_train and new_test should equal length of old_train"
+    if new_test.shape[1] != 12:
+        raise AssertionError("Public test set should have 12 columns")
+    if new_train.shape[1] != 13:
+        raise AssertionError("Public train set should have 13 columns")
+    if len(new_train) + len(new_test) != len(old_train):
+        raise AssertionError("Length of new_train and new_test should equal length of old_train")


 if __name__ == "__main__":
@@ -320,7 +320,8 @@ class DataScienceRDLoop(RDLoop):
            # only clean current workspace without affecting other loops.
            for k in "direct_exp_gen", "coding", "running":
                if k in prev_out and prev_out[k] is not None:
-                    assert isinstance(prev_out[k], DSExperiment)
+                    if not isinstance(prev_out[k], DSExperiment):
+                        raise TypeError(f"prev_out[{k!r}] must be an instance of DSExperiment")
                    clean_workspace(prev_out[k].experiment_workspace.workspace_path)

            # Backup the workspace (only necessary files are included)
@@ -213,7 +213,8 @@ class DSTrace(Trace[DataScienceScen, KnowledgeBase]):
        self, component: COMPONENT, search_list: list[tuple[DSExperiment, ExperimentFeedback]] = []
    ) -> bool:
        for exp, fb in search_list:
-            assert isinstance(exp.hypothesis, DSHypothesis), "Hypothesis should be DSHypothesis (and not None)"
+            if not isinstance(exp.hypothesis, DSHypothesis):
+                raise TypeError("Hypothesis should be DSHypothesis (and not None)")
            if exp.hypothesis.component == component and fb:
                return True
        return False
@@ -182,7 +182,7 @@ class ExpGen2Hypothesis(DSProposalV2ExpGen):

        success_fb_list = list(set(trace_fbs))
        logger.info(
-            f"Merge Hypothesis: select {len(success_fb_list)} from {len(trace_fbs)} SOTA experiments found in {len(leaves)} traces"
+            f"Merge Hypothesis: select {len(success_fb_list)} from {len(trace_fbs)} SOTA experiments found in {len(leaves)} traces"  # nosec B608 — not SQL, Bandit false positive on "select" in log message
        )

        if len(success_fb_list) > 0:
@@ -377,7 +377,8 @@ class ExpGen2TraceAndMergeV2(ExpGen):
            if DS_RD_SETTING.enable_multi_version_exp_gen:
                exp_gen_version_list = DS_RD_SETTING.exp_gen_version_list.split(",")
                for version in exp_gen_version_list:
-                    assert version in ["v3", "v2", "v1"]
+                    if version not in ["v3", "v2", "v1"]:
+                        raise ValueError(f"version must be 'v1', 'v2', or 'v3', got {version!r}")

                if len(trace.hist) == 0:
                    # set the proposal version for the first sub-trace
@@ -339,7 +339,8 @@ class DSProposalV1ExpGen(ExpGen):
            eda_output = sota_exp.experiment_workspace.file_dict.get("EDA.md", None)
        scenario_desc = trace.scen.get_scenario_all_desc(eda_output=eda_output)

-        assert sota_exp is not None, "SOTA experiment is not provided."
+        if sota_exp is None:
+            raise ValueError("SOTA experiment is not provided.")
        last_exp = trace.last_exp()
        # exp_and_feedback = trace.hist[-1]
        # last_exp = exp_and_feedback[0]
@@ -445,8 +446,10 @@ class DSProposalV1ExpGen(ExpGen):
                        json_target_type=dict[str, dict[str, str | dict] | str],
                    )
                )
-                assert "hypothesis_proposal" in resp_dict, "Hypothesis proposal not provided."
-                assert "task_design" in resp_dict, "Task design not provided."
+                if "hypothesis_proposal" not in resp_dict:
+                    raise ValueError("Hypothesis proposal not provided.")
+                if "task_design" not in resp_dict:
+                    raise ValueError("Task design not provided.")
                task_class = component_info["task_class"]
                hypothesis_proposal = resp_dict.get("hypothesis_proposal", {})
                hypothesis = DSHypothesis(
@@ -1149,8 +1152,10 @@ You help users retrieve relevant knowledge from community discussions and public
        )

        response_dict = json.loads(response)
-        assert response_dict.get("component") in HypothesisComponent.__members__, f"Invalid component"
-        assert response_dict.get("hypothesis") is not None, f"Invalid hypothesis"
+        if response_dict.get("component") not in HypothesisComponent.__members__:
+            raise ValueError(f"Invalid component: {response_dict.get('component')}")
+        if response_dict.get("hypothesis") is None:
+            raise ValueError("Invalid hypothesis")
        return response_dict

    # END: for support llm-based hypothesis selection  -----
@@ -1253,7 +1258,8 @@ You help users retrieve relevant knowledge from community discussions and public
            description=task_desc,
        )

-        assert isinstance(task, PipelineTask), f"Task {task_name} is not a PipelineTask, got {type(task)}"
+        if not isinstance(task, PipelineTask):
+            raise TypeError(f"Task {task_name} is not a PipelineTask, got {type(task)}")
        # only for llm with response schema.(TODO: support for non-schema llm?)
        # If the LLM provides a "packages" field (list[str]), compute runtime environment now and cache it for subsequent prompts in later loops.
        if isinstance(task_dict, dict) and "packages" in task_dict and isinstance(task_dict["packages"], list):
@@ -1,3 +1,4 @@
+import ast
 import json
 import os
 import pickle
@@ -292,7 +293,7 @@ class ValidationSelector(SOTAexpSelector):
        Sorts all valid experiments by score and returns the top N.
        """

-        mock_folder = f"/tmp/mock/{self.competition}"
+        mock_folder = f"/tmp/mock/{self.competition}"  # nosec B108 — Docker volume mount point derived from internal competition name

        try:
            data_py_code, grade_py_code = self._prepare_validation_scripts(
@@ -539,7 +540,7 @@ def process_experiment(

        # Run main script
        env = get_ds_env(
-            extra_volumes={f"/tmp/mock/{competition}/{input_folder}": input_folder},
+            extra_volumes={f"/tmp/mock/{competition}/{input_folder}": input_folder},  # nosec B108 — Docker volume mount point derived from internal competition name
            running_timeout_period=DS_RD_SETTING.full_timeout,
        )
        result = ws.run(env=env, entry="python main.py")
@@ -587,8 +588,8 @@ def _parsing_score(grade_stdout: str) -> Optional[float]:
        except:
            pass
        try:
-            # Priority 2: Eval dict
-            return float(eval(json_str)["score"])
+            # Priority 2: safe literal eval for Python-style dicts
+            return float(ast.literal_eval(json_str)["score"])
        except:
            pass
        try:
@@ -35,10 +35,11 @@ def select(X: pd.DataFrame) -> pd.DataFrame:
 class KGModelFeatureSelectionCoder(Developer[KGModelExperiment]):
    def develop(self, exp: KGModelExperiment) -> KGModelExperiment:
        target_model_type = exp.sub_tasks[0].model_type
-        assert target_model_type in KG_SELECT_MAPPING
+        if target_model_type not in KG_SELECT_MAPPING:
+            raise ValueError(f"target_model_type {target_model_type} not in KG_SELECT_MAPPING")
        if len(exp.experiment_workspace.data_description) == 1:
            code = (
-                Environment(undefined=StrictUndefined)
+                Environment(undefined=StrictUndefined)  # nosec B701 — renders Python code templates, not HTML; autoescape would corrupt code
                .from_string(DEFAULT_SELECTION_CODE)
                .render(feature_index_list=None)
            )
@@ -62,7 +63,7 @@ class KGModelFeatureSelectionCoder(Developer[KGModelExperiment]):
            chosen_index_to_list_index = [i - 1 for i in chosen_index]

            code = (
-                Environment(undefined=StrictUndefined)
+                Environment(undefined=StrictUndefined)  # nosec B701 — renders Python code templates, not HTML; autoescape would corrupt code
                .from_string(DEFAULT_SELECTION_CODE)
                .render(feature_index_list=chosen_index_to_list_index)
            )
@@ -165,7 +165,8 @@ class KGScenario(Scenario):
        return data_info

    def output_format(self, tag=None) -> str:
-        assert tag in [None, "feature", "model"]
+        if tag not in [None, "feature", "model"]:
+            raise ValueError(f"tag must be None, 'feature', or 'model', got {tag!r}")
        feature_output_format = f"""The feature code should output following the format:
 {T(".prompts:kg_feature_output_format").r()}"""
        model_output_format = f"""The model code should output following the format:\n""" + T(
@@ -180,7 +181,8 @@ class KGScenario(Scenario):
            return model_output_format

    def interface(self, tag=None) -> str:
-        assert tag in [None, "feature", "XGBoost", "RandomForest", "LightGBM", "NN"]
+        if tag not in [None, "feature", "XGBoost", "RandomForest", "LightGBM", "NN"]:
+            raise ValueError(f"tag must be None, 'feature', 'XGBoost', 'RandomForest', 'LightGBM', or 'NN', got {tag!r}")
        feature_interface = f"""The feature code should follow the interface:
 {T(".prompts:kg_feature_interface").r()}"""
        if tag == "feature":
@@ -195,7 +197,8 @@ class KGScenario(Scenario):
            return model_interface

    def simulator(self, tag=None) -> str:
-        assert tag in [None, "feature", "model"]
+        if tag not in [None, "feature", "model"]:
+            raise ValueError(f"tag must be None, 'feature', or 'model', got {tag!r}")

        kg_feature_simulator = (
            "The feature code will be sent to the simulator:\n" + T(".prompts:kg_feature_simulator").r()
@@ -79,12 +79,12 @@ def preprocess_script():
    This method applies the preprocessing steps to the training, validation, and test datasets.
    """
    if os.path.exists("/kaggle/input/X_train.pkl"):
-        X_train = pd.read_pickle("/kaggle/input/X_train.pkl")
-        X_valid = pd.read_pickle("/kaggle/input/X_valid.pkl")
-        y_train = pd.read_pickle("/kaggle/input/y_train.pkl")
-        y_valid = pd.read_pickle("/kaggle/input/y_valid.pkl")
-        X_test = pd.read_pickle("/kaggle/input/X_test.pkl")
-        others = pd.read_pickle("/kaggle/input/others.pkl")
+        X_train = pd.read_pickle("/kaggle/input/X_train.pkl")  # nosec B301 — trusted Kaggle input
+        X_valid = pd.read_pickle("/kaggle/input/X_valid.pkl")  # nosec B301
+        y_train = pd.read_pickle("/kaggle/input/y_train.pkl")  # nosec B301
+        y_valid = pd.read_pickle("/kaggle/input/y_valid.pkl")  # nosec B301
+        X_test = pd.read_pickle("/kaggle/input/X_test.pkl")  # nosec B301
+        others = pd.read_pickle("/kaggle/input/others.pkl")  # nosec B301
        y_train = pd.Series(y_train).reset_index(drop=True)
        y_valid = pd.Series(y_valid).reset_index(drop=True)

@@ -85,12 +85,12 @@ def preprocess_script():
    This method applies the preprocessing steps to the training, validation, and test datasets.
    """
    if os.path.exists("/kaggle/input/X_train.pkl"):
-        X_train = pd.read_pickle("/kaggle/input/X_train.pkl")
-        X_valid = pd.read_pickle("/kaggle/input/X_valid.pkl")
-        y_train = pd.read_pickle("/kaggle/input/y_train.pkl")
-        y_valid = pd.read_pickle("/kaggle/input/y_valid.pkl")
-        X_test = pd.read_pickle("/kaggle/input/X_test.pkl")
-        others = pd.read_pickle("/kaggle/input/others.pkl")
+        X_train = pd.read_pickle("/kaggle/input/X_train.pkl")  # nosec B301
+        X_valid = pd.read_pickle("/kaggle/input/X_valid.pkl")  # nosec B301
+        y_train = pd.read_pickle("/kaggle/input/y_train.pkl")  # nosec B301
+        y_valid = pd.read_pickle("/kaggle/input/y_valid.pkl")  # nosec B301
+        X_test = pd.read_pickle("/kaggle/input/X_test.pkl")  # nosec B301
+        others = pd.read_pickle("/kaggle/input/others.pkl")  # nosec B301

        return X_train, X_valid, y_train, y_valid, X_test, *others
    X_train, X_valid, y_train, y_valid = prepreprocess()
@@ -82,11 +82,11 @@ def preprocess_script():
    This method applies the preprocessing steps to the training, validation, and test datasets.
    """
    if os.path.exists("X_train.pkl"):
-        X_train = pd.read_pickle("X_train.pkl")
-        X_valid = pd.read_pickle("X_valid.pkl")
-        y_train = pd.read_pickle("y_train.pkl")
-        y_valid = pd.read_pickle("y_valid.pkl")
-        X_test = pd.read_pickle("X_test.pkl")
+        X_train = pd.read_pickle("X_train.pkl")  # nosec B301
+        X_valid = pd.read_pickle("X_valid.pkl")  # nosec B301
+        y_train = pd.read_pickle("y_train.pkl")  # nosec B301
+        y_valid = pd.read_pickle("y_valid.pkl")  # nosec B301
+        X_test = pd.read_pickle("X_test.pkl")  # nosec B301
        return X_train, X_valid, y_train, y_valid, X_test

    X_train, X_valid, y_train, y_valid, test, status_encoder, test_ids = prepreprocess()
@@ -73,12 +73,12 @@ def preprocess_script():
    This method applies the preprocessing steps to the training, validation, and test datasets.
    """
    if os.path.exists("/kaggle/input/X_train.pkl"):
-        X_train = pd.read_pickle("/kaggle/input/X_train.pkl")
-        X_valid = pd.read_pickle("/kaggle/input/X_valid.pkl")
-        y_train = pd.read_pickle("/kaggle/input/y_train.pkl")
-        y_valid = pd.read_pickle("/kaggle/input/y_valid.pkl")
-        X_test = pd.read_pickle("/kaggle/input/X_test.pkl")
-        others = pd.read_pickle("/kaggle/input/others.pkl")
+        X_train = pd.read_pickle("/kaggle/input/X_train.pkl")  # nosec B301
+        X_valid = pd.read_pickle("/kaggle/input/X_valid.pkl")  # nosec B301
+        y_train = pd.read_pickle("/kaggle/input/y_train.pkl")  # nosec B301
+        y_valid = pd.read_pickle("/kaggle/input/y_valid.pkl")  # nosec B301
+        X_test = pd.read_pickle("/kaggle/input/X_test.pkl")  # nosec B301
+        others = pd.read_pickle("/kaggle/input/others.pkl")  # nosec B301
        y_train = pd.Series(y_train).reset_index(drop=True)
        y_valid = pd.Series(y_valid).reset_index(drop=True)

@@ -87,7 +87,12 @@ def crawl_descriptions(
        content = e.get_attribute("innerHTML")
        contents.append(content)

-    assert len(subtitles) == len(contents) + 1 and subtitles[-1] == "Citation"
+    if not (len(subtitles) == len(contents) + 1 and subtitles[-1] == "Citation"):
+        raise AssertionError(
+            f"Expected len(contents)+1 == len(subtitles) and last subtitle == 'Citation', "
+            f"got len(subtitles)={len(subtitles)}, len(contents)={len(contents)}, "
+            f"last subtitle={subtitles[-1]!r}"
+        )
    for i in range(len(subtitles) - 1):
        descriptions[subtitles[i]] = contents[i]

@@ -307,7 +307,8 @@ class KGHypothesisGen(FactorAndModelHypothesisGen):
 class KGHypothesis2Experiment(FactorAndModelHypothesis2Experiment):
    def prepare_context(self, hypothesis: Hypothesis, trace: Trace) -> Tuple[dict, bool]:
        scenario = trace.scen.get_scenario_all_desc(filtered_tag="hypothesis_and_experiment")
-        assert isinstance(hypothesis, KGHypothesis)
+        if not isinstance(hypothesis, KGHypothesis):
+            raise TypeError("hypothesis must be an instance of KGHypothesis")
        experiment_output_format = (
            T("scenarios.kaggle.prompts:feature_experiment_output_format").r()
            if hypothesis.action in [KG_ACTION_FEATURE_ENGINEERING, KG_ACTION_FEATURE_PROCESSING]
@@ -1,5 +1,7 @@
+import logging
 import os
 from pathlib import Path
+
 """
 Qlib Factor Runner - Executes factor backtests in Docker.

@@ -9,15 +11,8 @@ NOTE: The @cache_with_pickle decorator was REMOVED from develop() because:
 - Docker-level caching (QlibDockerConf.enable_cache=False) is sufficient
 - The pickle cache caused 240+ factor generations but ZERO Docker backtests
 """
-from pathlib import Path
-from typing import Optional

 import pandas as pd
-from pandarallel import pandarallel
-
-
-pandarallel.initialize(verbose=1)
-
 from rdagent.app.qlib_rd_loop.conf import FactorBasePropSetting
 from rdagent.components.runner import CachedRunner
 from rdagent.core.exception import FactorEmptyError
@@ -29,6 +24,83 @@ from rdagent.scenarios.qlib.experiment.model_experiment import QlibModelExperime
 DIRNAME = Path(__file__).absolute().resolve().parent
 DIRNAME_local = Path.cwd()

+
+def _shift_daily_constant_factor_if_needed(factor_col: "pd.Series", factor_name: str) -> "pd.Series":
+    """Detect and fix look-ahead bias in daily-constant factors.
+
+    A factor is "daily-constant" when every minute bar within the same calendar
+    day carries an identical value. This happens when LLM code computes a daily
+    aggregate (e.g. today's log return) and forward-fills it across all intraday
+    bars without shifting — meaning the end-of-day value is visible at 00:00.
+
+    Fix: shift by one trading day so that the value assigned to day T is the
+    aggregate computed from day T-1, eliminating the forward-looking information.
+    """
+    import numpy as np
+
+    try:
+        notnull = factor_col.dropna()
+        if len(notnull) < 200:
+            return factor_col
+
+        datetimes = notnull.index.get_level_values("datetime")
+        dates = datetimes.normalize()
+
+        # Sample up to 50 random days and check intra-day uniqueness
+        unique_dates = pd.Series(dates.unique())
+        sample_dates = unique_dates.sample(min(50, len(unique_dates)), random_state=42)
+
+        daily_unique_counts = []
+        for d in sample_dates:
+            mask = dates == d
+            vals = notnull.values[mask]
+            if len(vals) > 1:
+                daily_unique_counts.append(len(np.unique(vals[~np.isnan(vals)])))
+
+        if not daily_unique_counts:
+            return factor_col
+
+        # If >90% of sampled days have exactly 1 unique value → daily-constant
+        fraction_constant = sum(1 for c in daily_unique_counts if c == 1) / len(daily_unique_counts)
+        if fraction_constant < 0.90:
+            return factor_col  # Intraday factor — no shift needed
+
+        logger.warning(
+            f"[LookAheadFix] Factor '{factor_name}' is daily-constant "
+            f"({fraction_constant:.0%} of days). Applying 1-day shift to remove look-ahead bias.",
+        )
+
+        # Shift: for each instrument, map daily values forward by 1 trading day
+        instruments = factor_col.index.get_level_values("instrument").unique()
+        shifted_parts = []
+        for inst in instruments:
+            inst_series = factor_col.xs(inst, level="instrument")
+            # Get one value per calendar day (the first non-null bar)
+            inst_dt = inst_series.index.normalize()
+            daily_vals = inst_series.groupby(inst_dt).first()
+            # Shift by 1 day
+            daily_vals_shifted = daily_vals.shift(1)
+            # Forward-fill back to minute bars
+            minute_idx = inst_series.index
+            minute_dates = minute_idx.normalize()
+            shifted_minute = minute_dates.map(daily_vals_shifted)
+            shifted_s = pd.Series(
+                shifted_minute.values,
+                index=pd.MultiIndex.from_arrays(
+                    [inst_series.index, [inst] * len(inst_series)],
+                    names=["datetime", "instrument"],
+                ),
+                name=factor_col.name,
+            )
+            shifted_parts.append(shifted_s)
+
+        return pd.concat(shifted_parts).sort_index()
+
+    except Exception as e:
+        logger.debug(f"[LookAheadFix] Could not apply daily shift for '{factor_name}': {e}")
+        return factor_col
+
+
 # TODO: supporting multiprocessing and keep previous results


@@ -43,13 +115,13 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
    """

    def calculate_information_coefficient(
-        self, concat_feature: pd.DataFrame, SOTA_feature_column_size: int, new_feature_columns_size: int
+        self, concat_feature: pd.DataFrame, SOTA_feature_column_size: int, new_feature_columns_size: int,
    ) -> pd.DataFrame:
        res = pd.Series(index=range(SOTA_feature_column_size * new_feature_columns_size))
        for col1 in range(SOTA_feature_column_size):
            for col2 in range(SOTA_feature_column_size, SOTA_feature_column_size + new_feature_columns_size):
                res.loc[col1 * new_feature_columns_size + col2 - SOTA_feature_column_size] = concat_feature.iloc[
-                    :, col1
+                    :, col1,
                ].corr(concat_feature.iloc[:, col2])
        return res

@@ -58,16 +130,21 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        # if the IC is larger than a threshold, remove the new_feature column
        # return the new_feature

+        from pandarallel import pandarallel
+        pandarallel.initialize(verbose=1)
+
        concat_feature = pd.concat([SOTA_feature, new_feature], axis=1)
        IC_max = (
            concat_feature.groupby("datetime")
            .parallel_apply(
-                lambda x: self.calculate_information_coefficient(x, SOTA_feature.shape[1], new_feature.shape[1])
+                lambda x: self.calculate_information_coefficient(x, SOTA_feature.shape[1], new_feature.shape[1]),
            )
            .mean()
        )
        IC_max.index = pd.MultiIndex.from_product([range(SOTA_feature.shape[1]), range(new_feature.shape[1])])
        IC_max = IC_max.unstack().max(axis=0)
+        if not hasattr(IC_max, "index"):
+            return new_feature
        return new_feature.iloc[:, IC_max[IC_max < 0.99].index]

    def develop(self, exp: QlibFactorExperiment) -> QlibFactorExperiment:
@@ -82,7 +159,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        self._ensure_results_dirs()

        if exp.based_experiments and exp.based_experiments[-1].result is None:
-            logger.info(f"Baseline experiment execution ...")
+            logger.info("Baseline experiment execution ...")
            exp.based_experiments[-1] = self.develop(exp.based_experiments[-1])

        fbps = FactorBasePropSetting()
@@ -106,11 +183,11 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                base_exp for base_exp in exp.based_experiments if isinstance(base_exp, QlibFactorExperiment)
            ]
            if len(sota_factor_experiments_list) > 1:
-                logger.info(f"SOTA factor processing ...")
+                logger.info("SOTA factor processing ...")
                SOTA_factor = process_factor_data(sota_factor_experiments_list)

            # Process the new factors data
-            logger.info(f"New factor processing ...")
+            logger.info("New factor processing ...")
            new_factors = process_factor_data(exp)

            if new_factors.empty:
@@ -121,7 +198,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                new_factors = self.deduplicate_new_factors(SOTA_factor, new_factors)
                if new_factors.empty:
                    raise FactorEmptyError(
-                        "The factors generated in this round are highly similar to the previous factors. Please change the direction for creating new factors."
+                        "The factors generated in this round are highly similar to the previous factors. Please change the direction for creating new factors.",
                    )
                combined_factors = pd.concat([SOTA_factor, new_factors], axis=1).dropna()
            else:
@@ -132,7 +209,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            combined_factors = combined_factors.loc[:, ~combined_factors.columns.duplicated(keep="last")]
            new_columns = pd.MultiIndex.from_product([["feature"], combined_factors.columns])
            combined_factors.columns = new_columns
-            logger.info(f"Factor data processing completed.")
+            logger.info("Factor data processing completed.")

            num_features = len(exp.base_features) + len(combined_factors.columns)

@@ -151,10 +228,10 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                    sota_model_exp = base_exp
                    exist_sota_model_exp = True
                    break
-            logger.info(f"Experiment execution ...")
+            logger.info("Experiment execution ...")
            if exist_sota_model_exp:
                exp.experiment_workspace.inject_files(
-                    **{"model.py": sota_model_exp.sub_workspace_list[0].file_dict["model.py"]}
+                    **{"model.py": sota_model_exp.sub_workspace_list[0].file_dict["model.py"]},
                )
                sota_training_hyperparameters = sota_model_exp.sub_tasks[0].training_hyperparameters
                if sota_training_hyperparameters:
@@ -165,19 +242,19 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                            "early_stop": str(sota_training_hyperparameters.get("early_stop", 10)),
                            "batch_size": str(sota_training_hyperparameters.get("batch_size", 256)),
                            "weight_decay": str(sota_training_hyperparameters.get("weight_decay", 0.0001)),
-                        }
+                        },
                    )
                sota_model_type = sota_model_exp.sub_tasks[0].model_type
                if sota_model_type == "TimeSeries":
                    env_to_use.update(
-                        {"dataset_cls": "TSDatasetH", "num_features": num_features, "step_len": 20, "num_timesteps": 20}
+                        {"dataset_cls": "TSDatasetH", "num_features": num_features, "step_len": 20, "num_timesteps": 20},
                    )
                elif sota_model_type == "Tabular":
                    env_to_use.update({"dataset_cls": "DatasetH", "num_features": num_features})

                # model + combined factors
                result, stdout = exp.experiment_workspace.execute(
-                    qlib_config_name="conf_combined_factors_sota_model.yaml", run_env=env_to_use
+                    qlib_config_name="conf_combined_factors_sota_model.yaml", run_env=env_to_use,
                )
            else:
                # LGBM + combined factors
@@ -186,7 +263,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                    run_env=env_to_use,
                )
        else:
-            logger.info(f"Experiment execution ...")
+            logger.info("Experiment execution ...")
            if exp.base_feature_codes:
                factors = process_factor_data(exp)
                factors = factors.sort_index()
@@ -196,7 +273,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                target_path = exp.experiment_workspace.workspace_path / "combined_factors_df.parquet"
                # Save the combined factors to the workspace
                factors.to_parquet(target_path, engine="pyarrow")
-                logger.info(f"Factor data processing completed.")
+                logger.info("Factor data processing completed.")
                result, stdout = exp.experiment_workspace.execute(
                    qlib_config_name="conf_combined_factors.yaml",
                    run_env=env_to_use,
@@ -209,10 +286,10 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

        # Handle Qlib Docker backtest failure gracefully
        if result is None:
-            factor_name = getattr(exp.hypothesis, 'hypothesis', 'unknown')
+            factor_name = getattr(exp.hypothesis, "hypothesis", "unknown")
            logger.warning(
                f"Qlib Docker backtest returned None for '{factor_name}'. "
-                f"Attempting direct factor evaluation..."
+                f"Attempting direct factor evaluation...",
            )

            # Try to compute metrics directly from the factor's result.h5
@@ -224,7 +301,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            else:
                logger.error(
                    f"Both Qlib Docker backtest and direct evaluation failed for '{factor_name}'. "
-                    f"Skipping this factor and continuing."
+                    f"Skipping this factor and continuing.",
                )
                # Save failed run info for debugging
                self._save_failed_run(exp, stdout, error_type="result_none")
@@ -242,7 +319,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        if validation_result.get("has_issues"):
            logger.warning(
                f"Result validation warnings for factor '{getattr(exp.hypothesis, 'hypothesis', 'unknown')}': "
-                f"{validation_result['warnings']}"
+                f"{validation_result['warnings']}",
            )
            # Save warning info for debugging
            self._save_failed_run(exp, stdout, error_type="validation_warnings", validation=validation_result)
@@ -293,43 +370,43 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        details = {}

        factor_name = "unknown"
-        if hasattr(exp, 'hypothesis') and exp.hypothesis is not None:
-            factor_name = getattr(exp.hypothesis, 'hypothesis', 'unknown')
+        if hasattr(exp, "hypothesis") and exp.hypothesis is not None:
+            factor_name = getattr(exp.hypothesis, "hypothesis", "unknown")

        if isinstance(result, pd.Series):
            # Check IC
-            ic_value = result.get('IC', None)
-            details['ic_raw'] = ic_value
+            ic_value = result.get("IC", None)
+            details["ic_raw"] = ic_value
            if ic_value is None or (isinstance(ic_value, float) and (ic_value != ic_value)):  # NaN check
                warnings.append("IC is None/NaN — factor has no predictive power")
            else:
                try:
                    ic_float = float(ic_value)
-                    details['ic'] = ic_float
+                    details["ic"] = ic_float
                    if abs(ic_float) < 0.001:
                        warnings.append(
-                            f"IC is near zero ({ic_float:.6f}) — factor may not predict returns"
+                            f"IC is near zero ({ic_float:.6f}) — factor may not predict returns",
                        )
                except (ValueError, TypeError):
                    warnings.append(f"IC value is not numeric: {ic_value}")

            # Check positions (1day.pos)
-            pos_value = result.get('1day.pos', None)
-            details['positions_raw'] = pos_value
+            pos_value = result.get("1day.pos", None)
+            details["positions_raw"] = pos_value
            if pos_value is not None:
                try:
                    pos_float = float(pos_value)
-                    details['positions'] = pos_float
+                    details["positions"] = pos_float
                    if pos_float == 0:
                        warnings.append(
                            "1day.pos == 0 — model opened ZERO positions (stayed neutral). "
                            "Possible causes: (1) topk too high for single-asset, "
-                            "(2) signal threshold too restrictive, (3) no valid predictions"
+                            "(2) signal threshold too restrictive, (3) no valid predictions",
                        )
                    elif pos_float < 10:
                        warnings.append(
                            f"1day.pos = {pos_float:.0f} — very few positions opened. "
-                            f"Check signal threshold and topk settings"
+                            f"Check signal threshold and topk settings",
                        )
                except (ValueError, TypeError):
                    pass  # pos might be a string
@@ -337,24 +414,24 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            # Check if result is essentially empty (all values None or NaN)
            non_null_count = result.notna().sum()
            total_count = len(result)
-            details['non_null_metrics'] = int(non_null_count)
-            details['total_metrics'] = int(total_count)
+            details["non_null_metrics"] = int(non_null_count)
+            details["total_metrics"] = int(total_count)
            if non_null_count < 3:
                warnings.append(
                    f"Result has only {non_null_count}/{total_count} non-null metrics — "
-                    f"backtest likely produced empty results"
+                    f"backtest likely produced empty results",
                )

            # Check for key metrics
-            required_metrics = ['IC', '1day.excess_return_with_cost.shar', '1day.pos']
+            required_metrics = ["IC", "1day.excess_return_with_cost.shar", "1day.pos"]
            for metric_name in required_metrics:
                val = result.get(metric_name, None)
-                details[f'has_{metric_name}'] = val is not None
+                details[f"has_{metric_name}"] = val is not None

        elif isinstance(result, dict):
            # Dict-based result validation
-            ic_value = result.get('IC', result.get('ic', None))
-            details['ic_raw'] = ic_value
+            ic_value = result.get("IC", result.get("ic", None))
+            details["ic_raw"] = ic_value
            if ic_value is None:
                warnings.append("IC is None — factor has no predictive power")

@@ -364,7 +441,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            "details": details,
        }

-    def _evaluate_factor_directly(self, exp, stdout: str) -> Optional[pd.Series]:
+    def _evaluate_factor_directly(self, exp, stdout: str) -> pd.Series | None:
        """
        Evaluate factor directly from its result.h5 file when Qlib Docker fails.

@@ -391,8 +468,19 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        import numpy as np

        try:
-            # Get workspace path
-            workspace_path = exp.experiment_workspace.workspace_path
+            # Get workspace path — factor code and result.h5 live in sub_workspace_list[0],
+            # not in experiment_workspace (which is the Qlib template workspace).
+            workspace_path = None
+            if exp.sub_workspace_list:
+                for ws in exp.sub_workspace_list:
+                    if ws is not None and hasattr(ws, "workspace_path"):
+                        candidate = ws.workspace_path / "result.h5"
+                        if candidate.exists():
+                            workspace_path = ws.workspace_path
+                            break
+            if workspace_path is None:
+                # Fallback to experiment_workspace
+                workspace_path = exp.experiment_workspace.workspace_path
            if workspace_path is None:
                return None

@@ -409,6 +497,12 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            factor_col = factor_values.iloc[:, 0]
            factor_name = factor_values.columns[0]

+            # Detect and fix look-ahead bias in daily-constant factors.
+            # If a factor has the same value for all minute bars within each calendar day
+            # it was computed from same-day data (e.g. today's close return at 00:00).
+            # Fix: shift by 1 trading day so value at day T = aggregate of day T-1.
+            factor_col = _shift_daily_constant_factor_if_needed(factor_col, factor_name)
+
            # Load source data for forward returns
            data_path = (
                Path(__file__).parent.parent.parent.parent.parent
@@ -483,7 +577,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

            logger.info(
                f"Direct evaluation: IC={ic:.6f}, Sharpe={sharpe:.4f}, "
-                f"AnnRet={annualized_return:.4f}%, WR={win_rate:.2%}"
+                f"AnnRet={annualized_return:.4f}%, WR={win_rate:.2%}",
            )
            return result

@@ -492,7 +586,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            return None

    def _save_failed_run(self, exp, stdout: str, error_type: str = "unknown",
-                         validation: Optional[dict] = None) -> None:
+                         validation: dict | None = None) -> None:
        """
        Save failed run information to results/failed_runs.json for debugging.

@@ -519,20 +613,20 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

            # Get factor name
            factor_name = "unknown"
-            if hasattr(exp, 'hypothesis') and exp.hypothesis is not None:
-                factor_name = getattr(exp.hypothesis, 'hypothesis', 'unknown')
+            if hasattr(exp, "hypothesis") and exp.hypothesis is not None:
+                factor_name = getattr(exp.hypothesis, "hypothesis", "unknown")

            # Build failed run record
            failed_record = {
                "timestamp": datetime.now().isoformat(),
                "factor_name": factor_name,
                "error_type": error_type,
-                "stdout": stdout if stdout else "(empty)",
+                "stdout": stdout or "(empty)",
                "validation": validation,
                "experiment_details": {
-                    "base_features": list(getattr(exp, 'base_features', {}).keys()) if hasattr(exp, 'base_features') else [],
-                    "hypothesis": getattr(exp.hypothesis, 'hypothesis', str(getattr(exp, 'hypothesis', 'N/A')))
-                                  if hasattr(exp, 'hypothesis') else "N/A",
+                    "base_features": list(getattr(exp, "base_features", {}).keys()) if hasattr(exp, "base_features") else [],
+                    "hypothesis": getattr(exp.hypothesis, "hypothesis", str(getattr(exp, "hypothesis", "N/A")))
+                                  if hasattr(exp, "hypothesis") else "N/A",
                },
            }

@@ -555,11 +649,11 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

            failed_file.write_text(
                json.dumps(existing_records, indent=2, default=str, ensure_ascii=False),
-                encoding="utf-8"
+                encoding="utf-8",
            )
            logger.info(
                f"Failed run saved: {factor_name} (type={error_type}) "
-                f"→ {failed_file}"
+                f"→ {failed_file}",
            )

        except Exception as e:
@@ -582,21 +676,23 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            containing metric names like 'IC', '1day.excess_return_with_cost.shar', etc.
        """
        try:
-            import json
-            import pandas as pd
            from pathlib import Path
+
+            import pandas as pd
            from rdagent.components.backtesting import ResultsDatabase

-            # Get factor name from hypothesis
+            # Get factor name: prefer hypothesis, fallback to result Series 'factor_name' key
            factor_name = "unknown"
-            if hasattr(exp, 'hypothesis') and exp.hypothesis is not None:
-                factor_name = getattr(exp.hypothesis, 'hypothesis', 'unknown')
+            if hasattr(exp, "hypothesis") and exp.hypothesis is not None:
+                factor_name = getattr(exp.hypothesis, "hypothesis", "unknown")
+            if factor_name == "unknown" and isinstance(result, pd.Series) and "factor_name" in result.index:
+                factor_name = str(result["factor_name"])

            # Check if already rejected by protection
-            if getattr(exp, 'rejected_by_protection', False):
+            if getattr(exp, "rejected_by_protection", False):
                logger.info(
                    f"Factor rejected by protection, skipping DB save: "
-                    f"{getattr(exp, 'protection_reason', 'unknown')}"
+                    f"{getattr(exp, 'protection_reason', 'unknown')}",
                )
                return

@@ -612,47 +708,47 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            # Extract metrics from result (pd.Series from qlib_res.csv)
            metrics = {}
            if isinstance(result, pd.Series):
-                metrics['ic'] = self._safe_float(result.get('IC', None))
-                metrics['sharpe_ratio'] = self._safe_float(
-                    result.get('1day.excess_return_with_cost.shar',
-                    result.get('1day.excess_return_with_cost.sharpe', None))
+                metrics["ic"] = self._safe_float(result.get("IC", None))
+                metrics["sharpe_ratio"] = self._safe_float(
+                    result.get("1day.excess_return_with_cost.shar",
+                    result.get("1day.excess_return_with_cost.sharpe", None)),
                )
-                metrics['annualized_return'] = self._safe_float(
-                    result.get('1day.excess_return_with_cost.annualized_return', None)
+                metrics["annualized_return"] = self._safe_float(
+                    result.get("1day.excess_return_with_cost.annualized_return", None),
                )
-                metrics['max_drawdown'] = self._safe_float(
-                    result.get('1day.excess_return_with_cost.max_drawdown', None)
+                metrics["max_drawdown"] = self._safe_float(
+                    result.get("1day.excess_return_with_cost.max_drawdown", None),
                )
-                metrics['win_rate'] = self._safe_float(result.get('win_rate', None))
-                metrics['information_ratio'] = self._safe_float(
-                    result.get('1day.excess_return_with_cost.information_ratio', None)
+                metrics["win_rate"] = self._safe_float(result.get("win_rate", None))
+                metrics["information_ratio"] = self._safe_float(
+                    result.get("1day.excess_return_with_cost.information_ratio", None),
                )
-                metrics['volatility'] = self._safe_float(
-                    result.get('1day.excess_return_with_cost.std',
-                    result.get('1day.excess_return_with_cost.volatility', None))
+                metrics["volatility"] = self._safe_float(
+                    result.get("1day.excess_return_with_cost.std",
+                    result.get("1day.excess_return_with_cost.volatility", None)),
                )
                # Store raw metrics for JSON export
-                metrics['raw_metrics'] = result.to_dict()
+                metrics["raw_metrics"] = result.to_dict()
            elif isinstance(result, dict):
-                metrics['ic'] = self._safe_float(result.get('IC', result.get('ic', None)))
-                metrics['sharpe_ratio'] = self._safe_float(
-                    result.get('sharpe', result.get('sharpe_ratio', None))
+                metrics["ic"] = self._safe_float(result.get("IC", result.get("ic", None)))
+                metrics["sharpe_ratio"] = self._safe_float(
+                    result.get("sharpe", result.get("sharpe_ratio", None)),
                )
-                metrics['annualized_return'] = self._safe_float(result.get('annualized_return', None))
-                metrics['max_drawdown'] = self._safe_float(result.get('max_drawdown', None))
-                metrics['win_rate'] = self._safe_float(result.get('win_rate', None))
-                metrics['information_ratio'] = None
-                metrics['volatility'] = None
-                metrics['raw_metrics'] = result
+                metrics["annualized_return"] = self._safe_float(result.get("annualized_return", None))
+                metrics["max_drawdown"] = self._safe_float(result.get("max_drawdown", None))
+                metrics["win_rate"] = self._safe_float(result.get("win_rate", None))
+                metrics["information_ratio"] = None
+                metrics["volatility"] = None
+                metrics["raw_metrics"] = result

            # Result validation before saving (warnings, not blocking)
            self._log_result_warnings(factor_name, result, metrics)

            # Only save if we have at least IC or Sharpe
-            if metrics.get('ic') is None and metrics.get('sharpe_ratio') is None:
+            if metrics.get("ic") is None and metrics.get("sharpe_ratio") is None:
                logger.warning(
                    f"No valid IC/Sharpe for factor '{factor_name}', skipping DB save. "
-                    f"IC={metrics.get('ic')}, Sharpe={metrics.get('sharpe_ratio')}"
+                    f"IC={metrics.get('ic')}, Sharpe={metrics.get('sharpe_ratio')}",
                )
                return

@@ -663,19 +759,19 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            db_file = db_path / "backtest_results.db"

            # Parallel run isolation: use run-specific subdirectory if PARALLEL_RUN_ID is set
-            run_id = os.getenv("PARALLEL_RUN_ID", "0")
-            if run_id != "0":
+            parallel_run_id = os.getenv("PARALLEL_RUN_ID", "0")
+            if parallel_run_id != "0":
                # For parallel runs, save to isolated results directory
-                isolated_db_path = project_root / "results" / "runs" / f"run{run_id}" / "db"
+                isolated_db_path = project_root / "results" / "runs" / f"run{parallel_run_id}" / "db"
                isolated_db_path.mkdir(parents=True, exist_ok=True)
                db_file = isolated_db_path / "backtest_results.db"

            # Save to database
            db = ResultsDatabase(db_path=str(db_file))
-            run_id = db.add_backtest(factor_name=factor_name[:100], metrics=metrics)
+            db_run_id = db.add_backtest(factor_name=factor_name[:100], metrics=metrics)
            logger.info(
                f"Factor result saved to DB: {factor_name[:60]} "
-                f"(IC={metrics.get('ic')}, Sharpe={metrics.get('sharpe_ratio')}, run_id={run_id})"
+                f"(IC={metrics.get('ic')}, Sharpe={metrics.get('sharpe_ratio')}, run_id={db_run_id})"
            )

            # Extract factor code and description from experiment
@@ -683,10 +779,10 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

            # Also write a JSON summary to results/factors/ for file-based access
            self._save_factor_json(
-                factor_name, metrics, run_id,
+                factor_name, metrics, db_run_id,
                factor_code=factor_code,
                factor_description=factor_description,
-                exp=exp
+                exp=exp,
            )

            # Save factor values as parquet for strategy building
@@ -698,7 +794,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            import traceback
            logger.error(
                f"Database save failed for factor '{getattr(exp.hypothesis, 'hypothesis', 'unknown')}': {e}\n"
-                f"Traceback: {traceback.format_exc()}"
+                f"Traceback: {traceback.format_exc()}",
            )

    def _save_factor_json(self, factor_name: str, metrics: dict, run_id: int,
@@ -809,14 +905,14 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                factor_description = match.group(1).strip()[:500]
            else:
                # Try comments
-                lines = factor_code.split('\n')
+                lines = factor_code.split("\n")
                desc_lines = []
                for line in lines[:20]:
                    stripped = line.strip()
-                    if stripped.startswith('#') and not stripped.startswith('#!'):
+                    if stripped.startswith("#") and not stripped.startswith("#!"):
                        desc_lines.append(stripped[1:].strip())
                if desc_lines:
-                    factor_description = ' '.join(desc_lines)[:500]
+                    factor_description = " ".join(desc_lines)[:500]

        return factor_code, factor_description

@@ -824,41 +920,80 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        """
        Save factor time-series values as parquet for strategy building.

-        This is essential for walk-forward validation and strategy combination.
-
-        Parameters
-        ----------
-        factor_name : str
-            Name of the factor
-        exp : QlibFactorExperiment
-            The experiment with factor values
+        Reruns the factor code on the FULL 6-year dataset so the parquet covers
+        the complete backtest range (not just the debug 2024 subset).
        """
        import os as _os
+        import shutil
+        import subprocess
+        import sys
+        import tempfile

        try:
-            # Get workspace path
-            workspace_path = exp.experiment_workspace.workspace_path
+            # factor.py lives in sub_workspace_list[0], not experiment_workspace
+            workspace_path = None
+            if exp.sub_workspace_list:
+                for ws in exp.sub_workspace_list:
+                    if ws is not None and hasattr(ws, "workspace_path"):
+                        fp = ws.workspace_path / "factor.py"
+                        if fp.exists():
+                            workspace_path = ws.workspace_path
+                            break
+            if workspace_path is None:
+                workspace_path = exp.experiment_workspace.workspace_path
            if workspace_path is None:
                return

-            result_h5 = workspace_path / "result.h5"
-            if not result_h5.exists():
+            factor_py = workspace_path / "factor.py"
+            if not factor_py.exists():
                return

-            # Read factor values
+            project_root = Path(__file__).parent.parent.parent.parent.parent
+            full_data = (
+                project_root
+                / "git_ignore_folder"
+                / "factor_implementation_source_data"
+                / "intraday_pv.h5"
+            )
+            if not full_data.exists():
+                return
+
+            # Run factor code on full data in a temp workspace
            import pandas as pd
-            df = pd.read_hdf(str(result_h5), key="data")
+            with tempfile.TemporaryDirectory(prefix="predix_fullval_") as tmp_dir:
+                tmp = Path(tmp_dir)
+                shutil.copy(str(factor_py), str(tmp / "factor.py"))
+                shutil.copy(str(full_data), str(tmp / "intraday_pv.h5"))
+
+                ret = subprocess.run(
+                    [sys.executable, "factor.py"],
+                    cwd=str(tmp),
+                    capture_output=True,
+                    timeout=300,
+                    check=False,
+                )
+                if ret.returncode != 0:
+                    logger.warning(
+                        f"Full-data factor run failed (exit {ret.returncode}): "
+                        f"{ret.stderr[:500] if ret.stderr else '(no stderr)'}"
+                    )
+                    # Fall back to debug-data result if full-data run fails
+                    result_h5 = workspace_path / "result.h5"
+                    if not result_h5.exists():
+                        return
+                    df = pd.read_hdf(str(result_h5), key="data")
+                else:
+                    result_h5_full = tmp / "result.h5"
+                    if not result_h5_full.exists():
+                        return
+                    df = pd.read_hdf(str(result_h5_full), key="data")
+
            if df is None or df.empty:
                return

-            # Get the factor series (first column)
            series = df.iloc[:, 0]
            series.name = factor_name

-            # Save to results/factors/values/
-            project_root = Path(__file__).parent.parent.parent.parent.parent
-
-            # Parallel run isolation
            parallel_run_id = _os.getenv("PARALLEL_RUN_ID", "0")
            if parallel_run_id != "0":
                values_dir = project_root / "results" / "runs" / f"run{parallel_run_id}" / "factors" / "values"
@@ -866,17 +1001,12 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                values_dir = project_root / "results" / "factors" / "values"

            values_dir.mkdir(parents=True, exist_ok=True)
-
-            # Safe filename
            safe_name = factor_name.replace("/", "_").replace("\\", "_").replace(" ", "_")[:100]
            parquet_path = values_dir / f"{safe_name}.parquet"
+            series.to_frame().to_parquet(str(parquet_path))

-            # Save as parquet (with datetime index)
-            series.to_parquet(str(parquet_path))
-
-        except Exception as e:
-            # Don't let factor value saving break the main workflow
-            pass
+        except Exception:
+            logging.debug("Error in save_factor_values_to_parquet", exc_info=True)

    def _log_result_warnings(self, factor_name: str, result, metrics: dict) -> None:
        """
@@ -897,7 +1027,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        warnings_list = []

        # Check IC
-        ic = metrics.get('ic')
+        ic = metrics.get("ic")
        if ic is None:
            warnings_list.append("IC is None — factor has no predictive power")
        elif abs(ic) < 0.001:
@@ -905,7 +1035,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

        # Check positions (1day.pos) — CRITICAL for EURUSD
        if isinstance(result, pd.Series):
-            pos_value = result.get('1day.pos', None)
+            pos_value = result.get("1day.pos", None)
            if pos_value is not None:
                try:
                    pos_float = float(pos_value)
@@ -913,23 +1043,23 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
                        warnings_list.append(
                            "WARNING: 1day.pos == 0 — ZERO positions opened! "
                            "Model stayed completely neutral. Check Qlib config: "
-                            "ensure topk=1 and market=eurusd for single-asset trading."
+                            "ensure topk=1 and market=eurusd for single-asset trading.",
                        )
                    elif pos_float < 10:
                        warnings_list.append(
                            f"Low position count: 1day.pos = {pos_float:.0f} — "
-                            f"model traded very rarely"
+                            f"model traded very rarely",
                        )
                except (ValueError, TypeError):
                    pass

        # Check Sharpe
-        sharpe = metrics.get('sharpe_ratio')
+        sharpe = metrics.get("sharpe_ratio")
        if sharpe is not None and abs(sharpe) < 0.1:
            warnings_list.append(f"Sharpe near zero ({sharpe:.4f}) — no risk-adjusted edge")

        # Check max drawdown
-        mdd = metrics.get('max_drawdown')
+        mdd = metrics.get("max_drawdown")
        if mdd is not None and mdd < -0.5:
            warnings_list.append(f"Extreme drawdown: {mdd:.2%} — high risk factor")

@@ -944,7 +1074,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            return None
        try:
            f = float(value)
-            if pd.isna(f) or f == float('inf') or f == float('-inf'):
+            if pd.isna(f) or f == float("inf") or f == float("-inf"):
                return None
            return f
        except (ValueError, TypeError):
@@ -987,7 +1117,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

        if protection_result.should_block:
            logger.warning(
-                f"Factor {factor_name} rejected by protection manager: {protection_result.reason}"
+                f"Factor {factor_name} rejected by protection manager: {protection_result.reason}",
            )
            # Mark factor as rejected by protection
            exp.rejected_by_protection = True
@@ -1012,8 +1142,8 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
        from pathlib import Path

        factor_name = "unknown"
-        if hasattr(exp, 'hypothesis') and exp.hypothesis is not None:
-            factor_name = getattr(exp.hypothesis, 'hypothesis', 'unknown')
+        if hasattr(exp, "hypothesis") and exp.hypothesis is not None:
+            factor_name = getattr(exp.hypothesis, "hypothesis", "unknown")

        # Build log entry
        log_entry = {
@@ -1025,42 +1155,42 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):
            "annualized_return": None,
            "max_drawdown": None,
            "win_rate": None,
-            "rejected_by_protection": getattr(exp, 'rejected_by_protection', False),
-            "protection_reason": getattr(exp, 'protection_reason', None),
+            "rejected_by_protection": getattr(exp, "rejected_by_protection", False),
+            "protection_reason": getattr(exp, "protection_reason", None),
        }

        # Extract metrics if available
        if result is not None:
-            if hasattr(result, 'get'):  # pd.Series or dict
-                ic_val = result.get('IC', result.get('ic', None))
-                log_entry['ic'] = self._safe_float(ic_val) if ic_val is not None else None
+            if hasattr(result, "get"):  # pd.Series or dict
+                ic_val = result.get("IC", result.get("ic", None))
+                log_entry["ic"] = self._safe_float(ic_val) if ic_val is not None else None

-                sharpe_val = result.get('1day.excess_return_with_cost.shar',
-                                        result.get('1day.excess_return_with_cost.sharpe',
-                                        result.get('sharpe', None)))
-                log_entry['sharpe'] = self._safe_float(sharpe_val) if sharpe_val is not None else None
+                sharpe_val = result.get("1day.excess_return_with_cost.shar",
+                                        result.get("1day.excess_return_with_cost.sharpe",
+                                        result.get("sharpe", None)))
+                log_entry["sharpe"] = self._safe_float(sharpe_val) if sharpe_val is not None else None

-                ann_ret = result.get('1day.excess_return_with_cost.annualized_return',
-                                     result.get('annualized_return', None))
-                log_entry['annualized_return'] = self._safe_float(ann_ret) if ann_ret is not None else None
+                ann_ret = result.get("1day.excess_return_with_cost.annualized_return",
+                                     result.get("annualized_return", None))
+                log_entry["annualized_return"] = self._safe_float(ann_ret) if ann_ret is not None else None

-                mdd = result.get('1day.excess_return_with_cost.max_drawdown',
-                                 result.get('max_drawdown', None))
-                log_entry['max_drawdown'] = self._safe_float(mdd) if mdd is not None else None
+                mdd = result.get("1day.excess_return_with_cost.max_drawdown",
+                                 result.get("max_drawdown", None))
+                log_entry["max_drawdown"] = self._safe_float(mdd) if mdd is not None else None

-                wr = result.get('win_rate', None)
-                log_entry['win_rate'] = self._safe_float(wr) if wr is not None else None
+                wr = result.get("win_rate", None)
+                log_entry["win_rate"] = self._safe_float(wr) if wr is not None else None

            # Determine status
-            if log_entry['ic'] is not None or log_entry['sharpe'] is not None:
-                log_entry['status'] = "success"
-            elif getattr(exp, 'rejected_by_protection', False):
-                log_entry['status'] = "rejected_protection"
+            if log_entry["ic"] is not None or log_entry["sharpe"] is not None:
+                log_entry["status"] = "success"
+            elif getattr(exp, "rejected_by_protection", False):
+                log_entry["status"] = "rejected_protection"
            else:
-                log_entry['status'] = "no_valid_metrics"
+                log_entry["status"] = "no_valid_metrics"
        else:
-            log_entry['status'] = "execution_failed"
-            log_entry['reason'] = "Result was None"
+            log_entry["status"] = "execution_failed"
+            log_entry["reason"] = "Result was None"

        # Write to results/logs/
        try:
@@ -1083,7 +1213,7 @@ class QlibFactorRunner(CachedRunner[QlibFactorExperiment]):

            logger.info(
                f"Run log written for '{factor_name[:50]}': "
-                f"status={log_entry['status']}, IC={log_entry['ic']}, Sharpe={log_entry['sharpe']}"
+                f"status={log_entry['status']}, IC={log_entry['ic']}, Sharpe={log_entry['sharpe']}",
            )
        except Exception as e:
            logger.error(f"Failed to write run log: {e}")
@@ -203,12 +203,14 @@ class QlibModelRunner(CachedRunner[QlibModelExperiment]):

            # Save to database
            db = ResultsDatabase()
-            run_id = db.add_backtest(factor_name=factor_name[:100], metrics=metrics)
-            logger.info(
-                f"Model result saved to DB: {factor_name[:50]} "
-                f"(IC={metrics.get('ic')}, Sharpe={metrics.get('sharpe_ratio')}, run_id={run_id})"
-            )
-            db.close()
+            try:
+                run_id = db.add_backtest(factor_name=factor_name[:100], metrics=metrics)
+                logger.info(
+                    f"Model result saved to DB: {factor_name[:50]} "
+                    f"(IC={metrics.get('ic')}, Sharpe={metrics.get('sharpe_ratio')}, run_id={run_id})"
+                )
+            finally:
+                db.close()

        except Exception as e:
            logger.warning(f"Database save failed for model {getattr(exp.hypothesis, 'hypothesis', 'unknown')}: {e}")
@@ -241,6 +241,7 @@ class StrategyBuilder:
                if data.get("status") == "success" and data.get("ic") is not None:
                    factors.append(data)
            except Exception:
+                logger.warning("Failed to load factor file %s", f, exc_info=True)
                continue

        # Sort by absolute IC
@@ -30,7 +30,8 @@ def _build_execute_calls(exp: QlibFactorExperiment, base_feature_workspaces: lis
    execute_calls = []

    if exp.sub_tasks:
-        assert isinstance(exp.prop_dev_feedback, CoSTEERMultiFeedback)
+        if not isinstance(exp.prop_dev_feedback, CoSTEERMultiFeedback):
+            raise TypeError("exp.prop_dev_feedback must be of type CoSTEERMultiFeedback")
        execute_calls.extend(
            (implementation.execute, ("All",))
            for implementation, feedback in zip(exp.sub_workspace_list, exp.prop_dev_feedback)
@@ -23,14 +23,25 @@ $low: low price at 1-minute bar.
 $volume: volume at 1-minute bar (tick volume for FX).

 ## Important Notes for 1min Data
- 96 bars = 1 trading day (24 hours for FX)
+- 1 bar = 1 minute (confirmed)
 - 16 bars = 16 minutes
- 4 bars = 4 minutes
- 1 bar = 1 minute
+- 60 bars = 1 hour
+- ~1440 bars = 1 full trading day (FX trades nearly 24h, Mon 00:00 - Fri 22:00 UTC approx.)
+- Typical bars per calendar day: ~1200-1440 (varies by weekday, holidays have fewer)
+- Do NOT assume 96 bars/day — the actual count depends on the date
 - Data range: 2020-01-01 to 2026-03-20
 - Instrument: EURUSD
 - Timezone: UTC

+## IMPORTANT: Bars per Day Correction
+The dataset has approximately 1440 bars per full trading day (1 bar = 1 minute, ~24h of FX trading).
+Some older documentation incorrectly stated "96 bars = 1 day" — this is WRONG. Always use:
+- 60 bars = 1 hour
+- 480 bars = 8 hours (London session 08:00-16:00 UTC)
+- 180 bars = 3 hours (London/NY overlap 13:00-16:00 UTC)
+Use datetime hour filtering (e.g., `df[df.index.get_level_values('datetime').hour.between(8, 15)]`)
+to select session bars — do NOT use bar-count offsets to define sessions.
+
 ## Session Times (UTC)
 - Asian: 00:00-08:00 UTC (low volatility)
 - London: 08:00-16:00 UTC (high volatility)
@@ -104,7 +104,7 @@ qlib_factor_strategy: |-
     result_df.columns = ['daily_volume_price_divergence']
     ```

-  4. **Process ALL data — do not filter dates**: The source HDF5 contains data from 2020-01-01 to 2026-03-20. Do NOT filter to a single year. If your output has only 314 entries (one year of daily data), the factor will be rejected. Expected output: ~1500+ daily entries for 2020-2026.
+  4. **Process ALL data — do not filter dates**: The source HDF5 contains data from 2020-01-01 to 2026-03-20 (development runs may use a 2024-only debug dataset with ~300 entries, which is acceptable). Do NOT filter to a single year in your code. Write your code to process whatever date range is available in the HDF5 file — do not hardcode date filters. Expected output for production data: ~1500+ daily entries for 2020-2026. Expected output for debug data: ~300 daily entries for 2024. Both are valid.

  5. **Use `transform()` instead of `apply()` for per-group calculations**: `transform()` preserves the original index while `apply()` may reduce the number of rows unexpectedly:
     ```python
@@ -121,6 +121,35 @@ qlib_factor_strategy: |-
     assert result_df.index.names == ['datetime', 'instrument'], f"Index names must be ['datetime', 'instrument'], got {result_df.index.names}"
     ```

+  7. **NEVER use same-day aggregations as the factor value — always shift by 1 day**: If your factor computes a daily aggregate (e.g. daily close return, daily OHLC range, daily volume), that aggregate is only known at end-of-day. Using it at the start of the same day is look-ahead bias. You MUST shift the daily aggregate by 1 day before forward-filling to minute bars:
+     ```python
+     # WRONG: look-ahead bias! Today's close return is not known at 00:00
+     daily_ret = df['$close'].groupby(level='instrument').resample('1D', level='datetime').last().pct_change()
+     result_df['my_factor'] = daily_ret.groupby(level='instrument').transform(lambda x: x.reindex(df.index.get_level_values('datetime'), method='ffill'))
+
+     # CORRECT: shift by 1 trading day so factor value at day T = aggregate of day T-1
+     daily_close = df.groupby([df.index.get_level_values('datetime').normalize(), df.index.get_level_values('instrument')])['$close'].last()
+     daily_close.index.names = ['date', 'instrument']
+     daily_ret = daily_close.groupby(level='instrument').pct_change().shift(1)  # <-- shift(1) is MANDATORY
+     # then map back to minute bars via ffill
+     ```
+     This rule applies to ALL daily aggregations: returns, OHLC stats, volume, momentum, slopes, etc.
+     **Session-based aggregations (London, NY, Asian session returns) are also daily aggregations** — the London
+     session (08:00-16:00 UTC) ends at 16:00, so its return must be shifted by 1 day before use.
+     Intraday rolling factors (e.g. 30-min rolling std computed at bar t using only bars t-N..t-1) do NOT need this shift.
+
+  8. **PREFER pure intraday rolling factors**: Factors that use only a trailing window of recent bars (e.g.
+     rolling(30).mean() of returns, RSI(14), Bollinger Band z-score) have NO look-ahead risk and vary every
+     minute. These are the best candidates for short-horizon (60-180 bar) prediction. Examples:
+     - Rolling 15-min / 30-min / 60-min return momentum (15, 30, 60 bars respectively)
+     - Rolling volatility (std of returns over 20-60 bars)
+     - Distance of close from N-bar moving average (z-score)
+     - RSI or similar oscillators computed on 1-min bars
+     - VWAP deviation (requires volume — use $volume column)
+     Always use `.shift(1)` on the lagged window (e.g. `rolling(N).mean().shift(1)`) to avoid using the
+     current bar's own price in its own feature value.
+     NOTE: 1 bar = 1 minute. The data has ~1440 bars per full trading day. Do NOT use 96 as a day proxy.
+
 qlib_factor_output_format: |-
  Your output should be a pandas dataframe similar to the following example information:
  <class 'pandas.core.frame.DataFrame'>
@@ -56,7 +56,8 @@ class QlibQuantScenario(Scenario):
        )

    def background(self, tag=None) -> str:
-        assert tag in [None, "factor", "model"]
+        if tag not in [None, "factor", "model"]:
+            raise ValueError(f"tag must be None, 'factor', or 'model', got {tag!r}")
        quant_background = "The background of the scenario is as follows:\n" + T(".prompts:qlib_quant_background").r(
            runtime_environment=self.get_runtime_environment(),
        )
@@ -83,7 +84,8 @@ class QlibQuantScenario(Scenario):
        return self._source_data

    def output_format(self, tag=None) -> str:
-        assert tag in [None, "factor", "model"]
+        if tag not in [None, "factor", "model"]:
+            raise ValueError(f"tag must be None, 'factor', or 'model', got {tag!r}")
        factor_output_format = (
            "The factor code should output the following format:\n" + T(".prompts:qlib_factor_output_format").r()
        )
@@ -99,7 +101,8 @@ class QlibQuantScenario(Scenario):
            return model_output_format

    def interface(self, tag=None) -> str:
-        assert tag in [None, "factor", "model"]
+        if tag not in [None, "factor", "model"]:
+            raise ValueError(f"tag must be None, 'factor', or 'model', got {tag!r}")
        factor_interface = (
            "The factor code should be written in the following interface:\n" + T(".prompts:qlib_factor_interface").r()
        )
@@ -115,7 +118,8 @@ class QlibQuantScenario(Scenario):
            return model_interface

    def simulator(self, tag=None) -> str:
-        assert tag in [None, "factor", "model"]
+        if tag not in [None, "factor", "model"]:
+            raise ValueError(f"tag must be None, 'factor', or 'model', got {tag!r}")
        factor_simulator = "The factor code will be sent to the simulator:\n" + T(".prompts:qlib_factor_simulator").r()
        model_simulator = "The model code will be sent to the simulator:\n" + T(".prompts:qlib_model_simulator").r()

@@ -185,7 +189,8 @@ class QlibQuantScenario(Scenario):
            return common_description(action) + interface(action) + output(action) + simulator(action)

    def get_runtime_environment(self, tag: str = None) -> str:
-        assert tag in [None, "factor", "model"]
+        if tag not in [None, "factor", "model"]:
+            raise ValueError(f"tag must be None, 'factor', or 'model', got {tag!r}")

        if tag is None or tag == "factor":
            # Use factor env to get the runtime environment
@@ -4,7 +4,7 @@ import shutil
 from pathlib import Path

 import pandas as pd
-from jinja2 import Environment, StrictUndefined
+from jinja2 import Environment, StrictUndefined, select_autoescape

 from rdagent.components.coder.factor_coder.config import FACTOR_COSTEER_SETTINGS
 from rdagent.utils.env import QTDockerEnv
@@ -21,14 +21,16 @@ def generate_data_folder_from_qlib():
        entry=f"python generate.py",
    )

-    assert (Path(__file__).parent / "factor_data_template" / "intraday_pv_all.h5").exists(), (
-        "intraday_pv_all.h5 is not generated. It means rdagent/scenarios/qlib/experiment/factor_data_template/generate.py is not executed correctly. Please check the log: \n"
-        + execute_log
-    )
-    assert (Path(__file__).parent / "factor_data_template" / "intraday_pv_debug.h5").exists(), (
-        "intraday_pv_debug.h5 is not generated. It means rdagent/scenarios/qlib/experiment/factor_data_template/generate.py is not executed correctly. Please check the log: \n"
-        + execute_log
-    )
+    if not (Path(__file__).parent / "factor_data_template" / "intraday_pv_all.h5").exists():
+        raise FileNotFoundError(
+            "intraday_pv_all.h5 is not generated. It means rdagent/scenarios/qlib/experiment/factor_data_template/generate.py is not executed correctly. Please check the log: \n"
+            + execute_log
+        )
+    if not (Path(__file__).parent / "factor_data_template" / "intraday_pv_debug.h5").exists():
+        raise FileNotFoundError(
+            "intraday_pv_debug.h5 is not generated. It means rdagent/scenarios/qlib/experiment/factor_data_template/generate.py is not executed correctly. Please check the log: \n"
+            + execute_log
+        )

    Path(FACTOR_COSTEER_SETTINGS.data_folder).mkdir(parents=True, exist_ok=True)
    shutil.copy(
@@ -67,7 +69,7 @@ def get_file_desc(p: Path, variable_list=[]) -> str:
    """
    p = Path(p)

-    JJ_TPL = Environment(undefined=StrictUndefined).from_string("""
+    JJ_TPL = Environment(undefined=StrictUndefined, autoescape=select_autoescape()).from_string("""
 # {{file_name}}

 ## File Type
@@ -1,4 +1,6 @@
+import logging
 import json
+import os
 from typing import List, Tuple

 from rdagent.components.coder.factor_coder.factor import FactorExperiment, FactorTask
@@ -9,6 +11,47 @@ from rdagent.scenarios.qlib.experiment.model_experiment import QlibModelExperime
 from rdagent.scenarios.qlib.experiment.quant_experiment import QlibQuantScenario
 from rdagent.utils.agent.tpl import T

+
+def _build_compressed_history(trace: Trace, max_history: int) -> str:
+    """Return hypothesis_and_feedback string with only `max_history` entries.
+
+    Older entries beyond the last 2 are compressed to one bullet line each.
+    """
+    if len(trace.hist) == 0:
+        return "No previous hypothesis and feedback available since it's the first round."
+
+    FULL_DETAIL = 2
+    old_hist = trace.hist[:-FULL_DETAIL] if len(trace.hist) > FULL_DETAIL else []
+    recent_hist = trace.hist[-FULL_DETAIL:] if len(trace.hist) > FULL_DETAIL else trace.hist
+
+    parts = []
+    if old_hist:
+        lines = ["## Earlier experiments (summarized):"]
+        for exp, fb in old_hist:
+            names = []
+            for task in exp.sub_tasks:
+                if task is not None and hasattr(task, "factor_name"):
+                    names.append(task.factor_name)
+                elif task is not None and hasattr(task, "model_type"):
+                    names.append(getattr(task, "model_type", "model"))
+            ic_str = ""
+            try:
+                if exp.result is not None and "IC" in exp.result.index:
+                    ic_str = f" IC={exp.result.loc['IC']:.4f}"
+            except Exception:
+                logging.debug("Exception caught", exc_info=True)
+            decision = "PASS" if fb.decision else "FAIL"
+            obs = (fb.observations or "")[:120].replace("\n", " ")
+            lines.append(f"- [{decision}]{ic_str} {', '.join(names) or 'unknown'}: {obs}")
+        parts.append("\n".join(lines))
+
+    if recent_hist:
+        rt = Trace(trace.scen)
+        rt.hist = recent_hist
+        parts.append(T("scenarios.qlib.prompts:hypothesis_and_feedback").r(trace=rt))
+
+    return "\n\n".join(parts)
+
 QlibFactorHypothesis = Hypothesis


@@ -17,13 +60,10 @@ class QlibFactorHypothesisGen(FactorHypothesisGen):
        super().__init__(scen)

    def prepare_context(self, trace: Trace) -> Tuple[dict, bool]:
-        hypothesis_and_feedback = (
-            T("scenarios.qlib.prompts:hypothesis_and_feedback").r(
-                trace=trace,
-            )
-            if len(trace.hist) > 0
-            else "No previous hypothesis and feedback available since it's the first round."
-        )
+        max_h = int(os.environ.get("QLIB_QUANT_MAX_FACTOR_HISTORY", "20"))
+        limited = Trace(trace.scen)
+        limited.hist = trace.hist[-max_h:] if len(trace.hist) > max_h else trace.hist
+        hypothesis_and_feedback = _build_compressed_history(limited, max_h)
        last_hypothesis_and_feedback = (
            T("scenarios.qlib.prompts:last_hypothesis_and_feedback").r(
                experiment=trace.hist[-1][0], feedback=trace.hist[-1][1]
@@ -70,15 +110,15 @@ class QlibFactorHypothesis2Experiment(FactorHypothesis2Experiment):
        if len(trace.hist) == 0:
            hypothesis_and_feedback = "No previous hypothesis and feedback available since it's the first round."
        else:
+            max_h = int(os.environ.get("QLIB_QUANT_MAX_FACTOR_HISTORY", "20"))
+            factor_hist = [
+                e for e in trace.hist
+                if not hasattr(e[0].hypothesis, "action") or e[0].hypothesis.action == "factor"
+            ][-max_h:]
            specific_trace = Trace(trace.scen)
-            for i in range(len(trace.hist) - 1, -1, -1):
-                if not hasattr(trace.hist[i][0].hypothesis, "action") or trace.hist[i][0].hypothesis.action == "factor":
-                    specific_trace.hist.insert(0, trace.hist[i])
-            if len(specific_trace.hist) > 0:
-                specific_trace.hist.reverse()
-                hypothesis_and_feedback = T("scenarios.qlib.prompts:hypothesis_and_feedback").r(
-                    trace=specific_trace,
-                )
+            specific_trace.hist = factor_hist
+            if specific_trace.hist:
+                hypothesis_and_feedback = _build_compressed_history(specific_trace, max_h)
            else:
                hypothesis_and_feedback = "No previous hypothesis and feedback available."

@@ -1,7 +1,7 @@
 import json
+import logging
 import os
 import random
-from typing import Tuple

 from rdagent.app.qlib_rd_loop.conf import QUANT_PROP_SETTING
 from rdagent.components.proposal import FactorAndModelHypothesisGen
@@ -41,7 +41,7 @@ class QlibQuantHypothesis(Hypothesis):
        action: str,
    ) -> None:
        super().__init__(
-            hypothesis, reason, concise_reason, concise_observation, concise_justification, concise_knowledge
+            hypothesis, reason, concise_reason, concise_observation, concise_justification, concise_knowledge,
        )
        self.action = action

@@ -53,10 +53,10 @@ Reason: {self.reason}


 class QlibQuantHypothesisGen(FactorAndModelHypothesisGen):
-    def __init__(self, scen: Scenario) -> Tuple[dict, bool]:
+    def __init__(self, scen: Scenario) -> None:
        super().__init__(scen)

-    def prepare_context(self, trace: Trace) -> Tuple[dict, bool]:
+    def prepare_context(self, trace: Trace) -> tuple[dict, bool]:

        # ========= Bandit ==========
        if QUANT_PROP_SETTING.action_selection == "bandit":
@@ -84,7 +84,7 @@ class QlibQuantHypothesisGen(FactorAndModelHypothesisGen):

            last_hypothesis_and_feedback = (
                T("scenarios.qlib.prompts:last_hypothesis_and_feedback").r(
-                    experiment=trace.hist[-1][0], feedback=trace.hist[-1][1]
+                    experiment=trace.hist[-1][0], feedback=trace.hist[-1][1],
                )
                if len(trace.hist) > 0
                else "No previous hypothesis and feedback available since it's the first round."
@@ -152,9 +152,41 @@ class QlibQuantHypothesisGen(FactorAndModelHypothesisGen):
                        factor_inserted = True
            if len(specific_trace.hist) > 0:
                specific_trace.hist.reverse()
-                hypothesis_and_feedback = T("scenarios.qlib.prompts:hypothesis_and_feedback").r(
-                    trace=specific_trace,
-                )
+                # Keep only the 2 most recent experiments in full detail; compress older ones
+                # to brief bullet points to stay within the LLM context window.
+                FULL_DETAIL_COUNT = 2
+                old_hist = specific_trace.hist[:-FULL_DETAIL_COUNT] if len(specific_trace.hist) > FULL_DETAIL_COUNT else []
+                recent_hist = specific_trace.hist[-FULL_DETAIL_COUNT:] if len(specific_trace.hist) > FULL_DETAIL_COUNT else specific_trace.hist
+
+                parts = []
+                if old_hist:
+                    summary_lines = ["## Earlier experiments (summarized):"]
+                    for exp, fb in old_hist:
+                        factor_names = []
+                        for task in exp.sub_tasks:
+                            if task is not None and hasattr(task, "factor_name"):
+                                factor_names.append(task.factor_name)
+                            elif task is not None and hasattr(task, "model_type"):
+                                factor_names.append(getattr(task, "model_type", "model"))
+                        names_str = ", ".join(factor_names) if factor_names else "unknown"
+                        ic_str = ""
+                        try:
+                            if exp.result is not None:
+                                ic_val = exp.result.loc["IC"] if "IC" in exp.result.index else ""
+                                ic_str = f" IC={ic_val:.4f}" if ic_val != "" else ""
+                        except Exception:
+                            logging.debug("Error getting IC", exc_info=True)
+                        decision_str = "PASS" if fb.decision else "FAIL"
+                        obs_short = (fb.observations or "")[:120].replace("\n", " ")
+                        summary_lines.append(f"- [{decision_str}]{ic_str} {names_str}: {obs_short}")
+                    parts.append("\n".join(summary_lines))
+
+                if recent_hist:
+                    recent_trace = Trace(specific_trace.scen)
+                    recent_trace.hist = recent_hist
+                    parts.append(T("scenarios.qlib.prompts:hypothesis_and_feedback").r(trace=recent_trace))
+
+                hypothesis_and_feedback = "\n\n".join(parts)
            else:
                hypothesis_and_feedback = "No previous hypothesis and feedback available."

@@ -162,7 +194,7 @@ class QlibQuantHypothesisGen(FactorAndModelHypothesisGen):
        for i in range(len(trace.hist) - 1, -1, -1):
            if trace.hist[i][0].hypothesis.action == action:
                last_hypothesis_and_feedback = T("scenarios.qlib.prompts:last_hypothesis_and_feedback").r(
-                    experiment=trace.hist[i][0], feedback=trace.hist[i][1]
+                    experiment=trace.hist[i][0], feedback=trace.hist[i][1],
                )
                break

@@ -171,7 +203,7 @@ class QlibQuantHypothesisGen(FactorAndModelHypothesisGen):
            for i in range(len(trace.hist) - 1, -1, -1):
                if trace.hist[i][0].hypothesis.action == "model" and trace.hist[i][1].decision is True:
                    sota_hypothesis_and_feedback = T("scenarios.qlib.prompts:sota_hypothesis_and_feedback").r(
-                        experiment=trace.hist[i][0], feedback=trace.hist[i][1]
+                        experiment=trace.hist[i][0], feedback=trace.hist[i][1],
                    )
                    break

@@ -77,6 +77,7 @@ def count_valid_factors() -> int:
            if data.get("status") == "success" and data.get("ic") is not None:
                count += 1
        except Exception:
+            logger.warning("Failed to load factor file %s", json_file, exc_info=True)
            continue

    return count
@@ -71,7 +71,7 @@ def submit_for_grading(grading_url: str, model_path: str) -> dict | None:
 def main():
    MODEL_PATH = os.environ.get("MODEL_PATH")
    DATA_PATH = os.environ.get("DATA_PATH")
-    OUTPUT_DIR = os.environ.get("OUTPUT_DIR", "/tmp/autorl_output")
+    OUTPUT_DIR = os.environ.get("OUTPUT_DIR", "/tmp/autorl_output")  # nosec B108 — Docker container output dir, configurable via env var
    GRADING_SERVER_URL = os.environ.get("GRADING_SERVER_URL", "")
    TRAIN_RATIO = float(os.environ.get("TRAIN_RATIO", "0.05"))
    NUM_EPOCHS = int(os.environ.get("NUM_EPOCHS", "3"))
@@ -391,7 +391,7 @@ def set_baseline():
    return jsonify({"baseline_score": score, "status": "set"})


-def run_server(task: str, base_model: str, workspace: str, host: str = "0.0.0.0", port: int = 5000):
+def run_server(task: str, base_model: str, workspace: str, host: str = "127.0.0.1", port: int = 5000):
    """启动服务器"""
    init_server(task, base_model, workspace)
    logger.info(f"Grading Server | task={task} | {host}:{port}")
@@ -435,7 +435,7 @@ class LocalServerContext(GradingServerContext):
        logger.info(f"[Local Mode] Starting evaluation server on port {self.port}...")
        self.server = init_server(self.task, self.base_model, self.workspace)

-        self._http_server = make_server("0.0.0.0", self.port, app, threaded=True)
+        self._http_server = make_server("0.0.0.0", self.port, app, threaded=True)  # nosec B104 — intentional: Docker sandbox requires all-interface binding
        self._thread = threading.Thread(target=self._http_server.serve_forever, daemon=True)
        self._thread.start()

@@ -488,7 +488,7 @@ if __name__ == "__main__":
    parser.add_argument("--base-model", type=str, default="")
    parser.add_argument("--workspace", type=str, default=".")
    parser.add_argument("--port", type=int, default=5000)
-    parser.add_argument("--host", type=str, default="0.0.0.0")
+    parser.add_argument("--host", type=str, default="127.0.0.1")
    args = parser.parse_args()

    run_server(args.task, args.base_model, args.workspace, args.host, args.port)
@@ -9,7 +9,7 @@ peft>=0.18.1

 # Evaluation
 opencompass==0.5.1
-setuptools<75  # uv venv doesn't include, opencompass depends on pkg_resources
+setuptools>=78.1.1  # Security fix: GHSA-8g6x-3r52-4m6c (path traversal in PackageIndex.download, arbitrary file write/RCE)

 # Inference acceleration (optional, TRL supports 0.10.2-0.12.0)
 # Security: Version >=0.14.0 fixes CVE-2026-22807 (RCE via auto_map dynamic module loading)
@@ -9,7 +9,7 @@ from pathlib import Path
 from typing import Any

 import yaml
-from jinja2 import Environment, FunctionLoader, StrictUndefined
+from jinja2 import Environment, FunctionLoader, StrictUndefined, select_autoescape

 from rdagent.core.conf import RD_AGENT_SETTINGS
 from rdagent.log import rdagent_logger as logger
@@ -38,7 +38,8 @@ def load_content(uri: str, caller_dir: Path | None = None, ftype: str = "yaml")
        caller_dir = get_caller_dir(upshift=1)
    # Parse the URI
    path_part, *yaml_trace = uri.split(":")
-    assert len(yaml_trace) <= 1, f"Invalid uri {uri}, only one yaml trace is allowed."
+    if len(yaml_trace) > 1:
+        raise ValueError(f"Invalid uri {uri}, only one yaml trace is allowed.")
    yaml_trace = [key for yt in yaml_trace for key in yt.split(".")]

    # load file_path with priorities.
@@ -126,7 +127,7 @@ class RDAT:
        # loader=FunctionLoader(load_conent) is for supporting grammar like below.
        # `{% include "scenarios.data_science.share:component_spec.DataLoadSpec" %}`
        rendered = (
-            Environment(undefined=StrictUndefined, loader=FunctionLoader(load_content))
+            Environment(undefined=StrictUndefined, loader=FunctionLoader(load_content), autoescape=select_autoescape())
            .from_string(self.template)
            .render(**context)
            .strip("\n")
@@ -436,13 +436,10 @@ class Env(Generic[ASpecificEnvConf]):
        else:
            timeout_cmd = f"timeout --kill-after=10 {self.conf.running_timeout_period} {entry}"
        entry_add_timeout = (
-            f"/bin/sh -c '"  # start of the sh command
-            + f"{timeout_cmd}; entry_exit_code=$?; "
+            "/bin/sh -c '"  # start of the sh command
+            + timeout_cmd.replace("'", "'\\''") + "; entry_exit_code=$?; "
            + (
                f"{_get_chmod_cmd(self.conf.mount_path)}; "
-                # We don't have to change the permission of the cache and input folder to remove it
-                # + f"if [ -d {self.conf.mount_path}/cache ]; then chmod 777 {self.conf.mount_path}/cache; fi; " +
-                #     f"if [ -d {self.conf.mount_path}/input ]; then chmod 777 {self.conf.mount_path}/input; fi; "
                if isinstance(self.conf, DockerConf)
                else ""
            )
@@ -614,13 +611,14 @@ class LocalEnv(Env[ASpecificLocalConf]):
        if self.conf.extra_volumes is not None:
            for lp, rp in self.conf.extra_volumes.items():
                volumes[lp] = rp["bind"] if isinstance(rp, dict) else rp
-            cache_path = "/tmp/sample" if "/sample/" in "".join(self.conf.extra_volumes.keys()) else "/tmp/full"
+            cache_path = "/tmp/sample" if "/sample/" in "".join(self.conf.extra_volumes.keys()) else "/tmp/full"  # nosec B108 — fixed Docker volume mount point, not a user-writable temp file
            Path(cache_path).mkdir(parents=True, exist_ok=True)
            volumes[cache_path] = T("scenarios.data_science.share:scen.cache_path").r()
        for lp, rp in running_extra_volume.items():
            volumes[lp] = rp

-        assert local_path is not None, "local_path should not be None"
+        if local_path is None:
+            raise ValueError("local_path should not be None")
        volumes = normalize_volumes(volumes, local_path)

        @contextlib.contextmanager
@@ -678,7 +676,7 @@ class LocalEnv(Env[ASpecificLocalConf]):
            cwd = Path(local_path).resolve() if local_path else None
            env = {k: str(v) if isinstance(v, int) else v for k, v in env.items()}

-            process = subprocess.Popen(
+            process = subprocess.Popen(  # nosec B602 — entry is an internal command string set by LocalEnvConf, not user input
                entry,
                cwd=cwd,
                env={**os.environ, **env},
@@ -761,12 +759,15 @@ class CondaConf(LocalConf):
        to ensure bin_path is set correctly even if the conda env was just created.
        """
        conda_path_result = subprocess.run(
-            f"conda run -n {self.conda_env_name} --no-capture-output env | grep '^PATH='",
+            ["conda", "run", "-n", self.conda_env_name, "--no-capture-output", "env"],
            capture_output=True,
            text=True,
-            shell=True,
        )
-        self.bin_path = conda_path_result.stdout.strip().split("=")[1] if conda_path_result.returncode == 0 else ""
+        if conda_path_result.returncode == 0:
+            path_lines = [l for l in conda_path_result.stdout.splitlines() if l.startswith("PATH=")]
+            self.bin_path = path_lines[0].split("=", 1)[1] if path_lines else ""
+        else:
+            self.bin_path = ""


 class MLECondaConf(CondaConf):
@@ -850,24 +851,22 @@ class QlibCondaEnv(LocalEnv[QlibCondaConf]):
    def prepare(self) -> None:
        """Prepare the conda environment if not already created."""
        try:
-            envs = subprocess.run("conda env list", capture_output=True, text=True, shell=True)
+            envs = subprocess.run(["conda", "env", "list"], capture_output=True, text=True)
            if self.conf.conda_env_name not in envs.stdout:
                print(f"[yellow]Conda env '{self.conf.conda_env_name}' not found, creating...[/yellow]")
                subprocess.check_call(
-                    f"conda create -y -n {self.conf.conda_env_name} python=3.10",
-                    shell=True,
+                    ["conda", "create", "-y", "-n", self.conf.conda_env_name, "python=3.10"],
                )
                subprocess.check_call(
-                    f"conda run -n {self.conf.conda_env_name} pip install --upgrade pip cython",
-                    shell=True,
+                    ["conda", "run", "-n", self.conf.conda_env_name, "pip", "install", "--upgrade", "pip", "cython"],
                )
                subprocess.check_call(
-                    f"conda run -n {self.conf.conda_env_name} pip install git+https://github.com/microsoft/qlib.git@2fb9380b342556ddb50a4b24e4fe8655d548b2b8",
-                    shell=True,
+                    ["conda", "run", "-n", self.conf.conda_env_name, "pip", "install",
+                     "git+https://github.com/microsoft/qlib.git@2fb9380b342556ddb50a4b24e4fe8655d548b2b8"],
                )
                subprocess.check_call(
-                    f"conda run -n {self.conf.conda_env_name} pip install catboost xgboost tables torch",
-                    shell=True,
+                    ["conda", "run", "-n", self.conf.conda_env_name, "pip", "install",
+                     "catboost", "xgboost", "tables", "torch"],
                )

        except Exception as e:
@@ -888,10 +887,9 @@ def _sync_conda_cache_with_real_envs() -> None:
    """Ensure the prepared cache includes environments that already exist on disk."""
    try:
        result = subprocess.run(
-            "conda env list",
+            ["conda", "env", "list"],
            capture_output=True,
            text=True,
-            shell=True,
            check=False,
        )
    except Exception as exc:  # pragma: no cover - best-effort helper
@@ -924,14 +922,19 @@ def _prepare_conda_env(env_name: str, requirements_file: Path, python_version: s
        python_version: Python version for the environment
    """
    # 1. Create conda environment if not exists
-    result = subprocess.run(f"conda env list | grep -q '^{env_name} '", shell=True)
-    if result.returncode != 0:
+    env_list = subprocess.run(["conda", "env", "list"], capture_output=True, text=True, check=False)
+    env_exists = any(
+        line.split()[0] == env_name
+        for line in env_list.stdout.splitlines()
+        if line and not line.startswith("#") and len(line.split()) > 0
+    )
+    if not env_exists:
        print(f"[yellow]Creating conda env '{env_name}' (Python {python_version})...[/yellow]")
-        subprocess.check_call(f"conda create -y -n {env_name} python={python_version}", shell=True)
-        subprocess.check_call(f"conda run -n {env_name} pip install --upgrade pip", shell=True)
+        subprocess.check_call(["conda", "create", "-y", "-n", env_name, f"python={python_version}"])
+        subprocess.check_call(["conda", "run", "-n", env_name, "pip", "install", "--upgrade", "pip"])

    print(f"[yellow]Installing dependencies from {requirements_file.name}...[/yellow]")
-    subprocess.check_call(f"conda run -n {env_name} pip install -r {requirements_file}", shell=True)
+    subprocess.check_call(["conda", "run", "-n", env_name, "pip", "install", "-r", str(requirements_file)])
    print(f"[green]Conda env '{env_name}' ready[/green]")

    _CONDA_ENV_PREPARED.add(env_name)
@@ -971,8 +974,8 @@ class FTCondaEnv(LocalEnv[FTCondaConf]):
            # Note: flash-attn>=2.8 is required for B200 (sm_100) support
            print("[yellow]Installing flash-attn (compiling, may take a few minutes)...[/yellow]")
            subprocess.check_call(
-                f"conda run -n {self.conf.conda_env_name} pip install 'flash-attn>=2.8' --no-build-isolation --no-cache-dir",
-                shell=True,
+                ["conda", "run", "-n", self.conf.conda_env_name, "pip", "install",
+                 "flash-attn>=2.8", "--no-build-isolation", "--no-cache-dir"],
            )

            # Re-update bin_path after prepare() in case the conda env was just created
@@ -1190,7 +1193,7 @@ class DockerEnv(Env[DockerConf]):
            with Progress(SpinnerColumn(), TextColumn("{task.description}")) as p:
                task = p.add_task("[cyan]Building image...")
                for part in resp_stream:
-                    lines = part.decode("utf-8").split("\r\n")
+                    lines = part.decode("utf-8", errors="replace").split("\r\n")
                    for line in lines:
                        if line.strip():
                            status_dict = json.loads(line)
@@ -1442,7 +1445,7 @@ class DockerEnv(Env[DockerConf]):
        if self.conf.extra_volumes is not None:
            for lp, rp in self.conf.extra_volumes.items():
                volumes[lp] = rp if isinstance(rp, dict) else {"bind": rp, "mode": self.conf.extra_volume_mode}
-            cache_path = "/tmp/sample" if "/sample/" in "".join(self.conf.extra_volumes.keys()) else "/tmp/full"
+            cache_path = "/tmp/sample" if "/sample/" in "".join(self.conf.extra_volumes.keys()) else "/tmp/full"  # nosec B108 — fixed Docker volume mount point, not a user-writable temp file
            Path(cache_path).mkdir(parents=True, exist_ok=True)
            volumes[cache_path] = {
                "bind": T("scenarios.data_science.share:scen.cache_path").r(),
@@ -1471,7 +1474,8 @@ class DockerEnv(Env[DockerConf]):
                cpu_count=self.conf.cpu_count,  # Set CPU limit
                **self._gpu_kwargs(client),
            )
-            assert container is not None  # Ensure container was created successfully
+            if container is None:
+                raise AssertionError("Docker container was not created successfully")
            logs = container.logs(stream=True)
            print(Rule("[bold green]Docker Logs Begin[/bold green]", style="dark_orange"))
            table = Table(title="Run Info", show_header=False)
@@ -1521,8 +1525,8 @@ class DockerEnv(Env[DockerConf]):
 class QTDockerEnv(DockerEnv):
    """Qlib Torch Docker"""

-    def __init__(self, conf: DockerConf = QlibDockerConf()):
-        super().__init__(conf)
+    def __init__(self, conf: DockerConf | None = None):
+        super().__init__(conf if conf is not None else QlibDockerConf())

    def prepare(self, *args, **kwargs) -> None:  # type: ignore[no-untyped-def]
        """
@@ -1541,15 +1545,15 @@ class QTDockerEnv(DockerEnv):
 class KGDockerEnv(DockerEnv):
    """Kaggle Competition Docker"""

-    def __init__(self, competition: str | None = None, conf: DockerConf = KGDockerConf()):
-        super().__init__(conf)
+    def __init__(self, competition: str | None = None, conf: DockerConf | None = None):
+        super().__init__(conf if conf is not None else KGDockerConf())


 class MLEBDockerEnv(DockerEnv):
    """MLEBench Docker"""

-    def __init__(self, conf: DockerConf = MLEBDockerConf()):
-        super().__init__(conf)
+    def __init__(self, conf: DockerConf | None = None):
+        super().__init__(conf if conf is not None else MLEBDockerConf())


 class FTDockerEnv(DockerEnv):
@@ -1565,8 +1569,8 @@ class FTDockerEnv(DockerEnv):
        export FT_DOCKER_save_logs_to_file=false # disable log file
    """

-    def __init__(self, conf: DockerConf = FTDockerConf()):
-        super().__init__(conf)
+    def __init__(self, conf: DockerConf | None = None):
+        super().__init__(conf if conf is not None else FTDockerConf())


 class BenchmarkDockerEnv(DockerEnv):
@@ -1583,5 +1587,5 @@ class BenchmarkDockerEnv(DockerEnv):
        export BENCHMARK_DOCKER_terminal_tail_lines=100  # show last 100 lines
    """

-    def __init__(self, conf: DockerConf = BenchmarkDockerConf()):
-        super().__init__(conf)
+    def __init__(self, conf: DockerConf | None = None):
+        super().__init__(conf if conf is not None else BenchmarkDockerConf())
@@ -15,19 +15,19 @@ import multiprocessing.queues
 import os
 import pickle
 from collections import defaultdict
+from collections.abc import Callable
 from dataclasses import dataclass
 from datetime import datetime, timezone
 from pathlib import Path
-from typing import Any, Callable, Optional, Union, cast
+from typing import Any, cast

 import psutil
-from tqdm.auto import tqdm
-
 from rdagent.core.conf import RD_AGENT_SETTINGS
 from rdagent.log import rdagent_logger as logger
 from rdagent.log.conf import LOG_SETTINGS
 from rdagent.log.timer import RD_Agent_TIMER_wrapper, RDAgentTimer
 from rdagent.utils.workflow.tracking import WorkflowTracker
+from tqdm.auto import tqdm


 class LoopMeta(type):
@@ -98,7 +98,7 @@ class LoopBase:
    skip_loop_error: tuple[type[BaseException], ...] = ()  # you can define a list of error that will skip current loop
    skip_loop_error_stepname: str | None = None  # if skip_loop_error exception happens, what's the next step to work on
    withdraw_loop_error: tuple[
-        type[BaseException], ...
+        type[BaseException], ...,
    ] = ()  # you can define a list of error that will withdraw current loop

    EXCEPTION_KEY = "_EXCEPTION"
@@ -129,8 +129,8 @@ class LoopBase:
        self.tracker = WorkflowTracker(self)  # Initialize tracker with this LoopBase instance

        # progress control
-        self.loop_n: Optional[int] = None  # remain loop count
-        self.step_n: Optional[int] = None  # remain step count
+        self.loop_n: int | None = None  # remain loop count
+        self.step_n: int | None = None  # remain step count

        self.semaphores: dict[str, asyncio.Semaphore] = {}

@@ -169,7 +169,7 @@ class LoopBase:
            self._pbar.close()
            del self._pbar

-    def _check_exit_conditions_on_step(self, loop_id: Optional[int] = None, step_id: Optional[int] = None) -> None:
+    def _check_exit_conditions_on_step(self, loop_id: int | None = None, step_id: int | None = None) -> None:
        """Check if the loop should continue or terminate.

        Raises
@@ -188,8 +188,7 @@ class LoopBase:
            if self.timer.is_timeout():
                logger.warning("Timeout, exiting the loop.")
                raise self.LoopTerminationError("Timer timeout")
-            else:
-                logger.info(f"Timer remaining time: {self.timer.remain_time()}")
+            logger.info(f"Timer remaining time: {self.timer.remain_time()}")

    async def _run_step(self, li: int, force_subproc: bool = False) -> None:
        """Execute a single step (next unrun step) in the workflow (async version with force_subproc option).
@@ -217,7 +216,7 @@ class LoopBase:

            with logger.tag(f"Loop_{li}.{name}"):
                start = datetime.now(timezone.utc)
-                func: Callable[..., Any] = cast(Callable[..., Any], getattr(self, name))
+                func: Callable[..., Any] = cast("Callable[..., Any]", getattr(self, name))

                next_step_idx = si + 1
                step_forward = True
@@ -233,15 +232,14 @@ class LoopBase:
                            # Using deepcopy is to avoid triggering errors like "RuntimeError: dictionary changed size during iteration"
                            # GUESS: Some content in self.loop_prev_out[li] may be in the middle of being changed.
                            result = await curr_loop.run_in_executor(
-                                pool, copy.deepcopy(func), copy.deepcopy(self.loop_prev_out[li])
+                                pool, copy.deepcopy(func), copy.deepcopy(self.loop_prev_out[li]),
                            )
+                    # auto determine whether to run async or sync
+                    elif asyncio.iscoroutinefunction(func):
+                        result = await func(self.loop_prev_out[li])
                    else:
-                        # auto determine whether to run async or sync
-                        if asyncio.iscoroutinefunction(func):
-                            result = await func(self.loop_prev_out[li])
-                        else:
-                            # Default: run sync function directly
-                            result = func(self.loop_prev_out[li])
+                        # Default: run sync function directly
+                        result = func(self.loop_prev_out[li])
                    # Store result in the nested dictionary
                    self.loop_prev_out[li][name] = result
                except Exception as e:
@@ -251,14 +249,13 @@ class LoopBase:
                            next_step_idx = self.steps.index(self.skip_loop_error_stepname)
                            if next_step_idx <= si:
                                raise RuntimeError(
-                                    f"Cannot skip backwards or to same step. Current: {si} ({name}), Target: {next_step_idx} ({self.skip_loop_error_stepname})"
+                                    f"Cannot skip backwards or to same step. Current: {si} ({name}), Target: {next_step_idx} ({self.skip_loop_error_stepname})",
                                ) from e
+                        # Default: jump to feedback step if exists, otherwise jump to the last step (record)
+                        elif "feedback" in self.steps:
+                            next_step_idx = self.steps.index("feedback")
                        else:
-                            # Default: jump to feedback step if exists, otherwise jump to the last step (record)
-                            if "feedback" in self.steps:
-                                next_step_idx = self.steps.index("feedback")
-                            else:
-                                next_step_idx = len(self.steps) - 1
+                            next_step_idx = len(self.steps) - 1
                        self.loop_prev_out[li][name] = None
                        self.loop_prev_out[li][self.EXCEPTION_KEY] = e
                    elif isinstance(e, self.withdraw_loop_error):
@@ -270,6 +267,11 @@ class LoopBase:
                        msg = "We have reset the loop instance, stop all the routines and resume."
                        raise self.LoopResumeError(msg) from e
                    else:
+                        # Do NOT advance step_idx for unhandled exceptions (e.g. LoopResumeError
+                        # propagating from _propose). Keeping step_idx at the current step lets
+                        # kickoff_loop retry step 0 on the next resume instead of permanently
+                        # corrupting the loop with a missing direct_exp_gen result.
+                        step_forward = False
                        raise  # re-raise unhandled exceptions
                finally:
                    # No matter the execution succeed or not, we have to finish the following steps
@@ -404,6 +406,8 @@ class LoopBase:
                self.close_pbar()

    def withdraw_loop(self, loop_idx: int) -> None:
+        if loop_idx <= 0:
+            raise RuntimeError(f"Cannot withdraw loop {loop_idx}: no previous loop exists.")
        prev_session_dir = self.session_folder / str(loop_idx - 1)
        prev_path = min(
            (p for p in prev_session_dir.glob("*_*") if p.is_file()),
@@ -496,7 +500,7 @@ class LoopBase:
            session_folder = path.parent.parent

        with path.open("rb") as f:
-            session = cast(LoopBase, pickle.load(f))
+            session = cast("LoopBase", pickle.load(f))

        # set session folder
        if checkout:
@@ -30,7 +30,8 @@ def wait_retry(
    >>> counter
    2
    """
-    assert retry_n > 0, "retry_n should be greater than 0"
+    if retry_n <= 0:
+        raise ValueError("retry_n should be greater than 0")

    def decorator(f: Callable[..., ASpecificRet]) -> Callable[..., ASpecificRet]:
        def wrapper(*args: Any, **kwargs: Any) -> ASpecificRet:
@@ -9,7 +9,6 @@ import datetime
 from typing import TYPE_CHECKING

 import pytz
-
 from rdagent.core.conf import RD_AGENT_SETTINGS
 from rdagent.log.timer import RD_Agent_TIMER_wrapper

@@ -84,12 +83,14 @@ class WorkflowTracker:
            # Log timer status if timer is started
            if self.loop_base.timer.started:
                remain_time = self.loop_base.timer.remain_time()
-                assert remain_time is not None
-                mlflow.log_metric("remain_time", remain_time.total_seconds())
-                mlflow.log_metric(
-                    "remain_percent",
-                    remain_time / self.loop_base.timer.all_duration * 100,
-                )
+                if remain_time is None:
+                    logger.warning("remain_time is None despite timer.started, skipping timer metrics")
+                else:
+                    mlflow.log_metric("remain_time", remain_time.total_seconds())
+                    mlflow.log_metric(
+                        "remain_percent",
+                        remain_time / self.loop_base.timer.all_duration * 100,
+                    )

        # Keep only the log_workflow_state method as it's the primary entry point now
        except Exception as e:
--- a/Show More
+++ b/Show More