diff --git a/rdagent/scenarios/data_science/dev/runner/eval.py b/rdagent/scenarios/data_science/dev/runner/eval.py
index 3ba63199..1b0fe1cf 100644
--- a/rdagent/scenarios/data_science/dev/runner/eval.py
+++ b/rdagent/scenarios/data_science/dev/runner/eval.py
@@ -133,7 +133,6 @@ class DSCoSTEERCoSTEEREvaluator(CoSTEEREvaluator):
             scenario=self.scen.get_scenario_all_desc(eda_output=implementation.file_dict.get("EDA.md", None)),
             is_sub_enabled=test_eval.is_sub_enabled(self.scen.competition),
             task_desc=target_task.get_task_information(),
-            runtime_environment=self.scen.get_runtime_environment(),
         )
         user_prompt = T(".prompts:DSCoSTEER_eval.user").r(
             code=implementation.all_codes,
diff --git a/rdagent/scenarios/data_science/dev/runner/prompts.yaml b/rdagent/scenarios/data_science/dev/runner/prompts.yaml
index aaf9ac72..60c2ebb3 100644
--- a/rdagent/scenarios/data_science/dev/runner/prompts.yaml
+++ b/rdagent/scenarios/data_science/dev/runner/prompts.yaml
@@ -19,10 +19,6 @@ DSCoSTEER_eval:
     The code is focusing on the following task
     {{ task_desc }}
 
-    ## Runtime Environment
-    You have following environment to run the code:
-    {{ runtime_environment }}
-
     ## Evaluation Guidelines
     1. Evaluate the code base based on several aspects, including execution correctness, return checking, and code quality.
     2. Ensure the code does not contain any incorrect, fabricated, or deceptive operations, such as mocking data, scores, or results.
diff --git a/rdagent/scenarios/data_science/scen/__init__.py b/rdagent/scenarios/data_science/scen/__init__.py
index 9c53bacf..3d90d07e 100644
--- a/rdagent/scenarios/data_science/scen/__init__.py
+++ b/rdagent/scenarios/data_science/scen/__init__.py
@@ -144,6 +144,7 @@ class DataScienceScen(Scenario):
             eda_output=None,
             sample_data_by_LLM=None,
             debug_time_limit=None,
+            runtime_environment=self.get_runtime_environment(),
         )
 
     def get_scenario_all_desc(self, eda_output=None) -> str:
@@ -162,6 +163,7 @@ class DataScienceScen(Scenario):
             eda_output=eda_output,
             sample_data_by_LLM=DS_RD_SETTING.sample_data_by_LLM,
             debug_time_limit=f"{DS_RD_SETTING.debug_timeout / 60 / 60 : .2f} hours",
+            runtime_environment=self.get_runtime_environment(),
         )
 
     def get_runtime_environment(self) -> str:
diff --git a/rdagent/scenarios/data_science/scen/prompts.yaml b/rdagent/scenarios/data_science/scen/prompts.yaml
index ec751939..a3f95995 100644
--- a/rdagent/scenarios/data_science/scen/prompts.yaml
+++ b/rdagent/scenarios/data_science/scen/prompts.yaml
@@ -43,6 +43,12 @@ scenario_description: |-
   Your are also required to include a debug mode in your code, the debug code's execution is limited to **{{ debug_time_limit }}**. You should make sure 10 percent of the data training one epoch can be finished within this time limit. If not, your should propose a new debug strategy in your task.
   {% endif %}{% endif %}
 
+  {% if runtime_environment is not none %}
+  ====== Runtime Environment ======
+  You have following environment to run the code:
+  {{ runtime_environment }}
+  {% endif %}
+
 competition_description_template:
   system: |-
     You are a data science assistant that extracts structured information from unstructured text.