From 5a892e6b0d4cce1d5618f4b063e21a913b81adae Mon Sep 17 00:00:00 2001 From: Theodore Song Date: Fri, 21 Aug 2026 15:18:49 -0400 Subject: [PATCH] Broaden agent-scoped forward learning --- README.md | 9 +++++++ index.html | 74 ++++++++++++++++++++++++++++++++++++++++++++---------- sw.js | 2 +- 3 files changed, 71 insertions(+), 14 deletions(-) diff --git a/README.md b/README.md index 28f6496..1202af1 100644 --- a/README.md +++ b/README.md @@ -337,6 +337,15 @@ independent events for per-agent 6-hour risk vetoes and 24-hour/72-hour promotion confidence checks without relaxing execution, cost, or evidence requirements. +Build 81 separates research eligibility from capital eligibility. Directional +trend and reversal candidates are tagged for every strategy whose broader +mandate would study them, including observation-only setups; actual positions +still require the strategy's stricter quality, edge, evidence, and forward +promotion gates. Sports-pilot and priced-bundle records are excluded from the +directional learner. Each agent report now audits the global pending queue as +legacy shared, strategy-tagged, unassigned, and tagged-for-this-agent counts so +learning differences are visible instead of inferred from a single total. + Build 77 separates the directional learner's evidence lineage from the global strategy release. Code-history verification found the same trend/reversal generator and 24-hour/72-hour grading policy in Strategy 51 through Strategy diff --git a/index.html b/index.html index 9de5818..970c4e6 100644 --- a/index.html +++ b/index.html @@ -341,7 +341,7 @@ footer{margin-top:34px;padding-top:22px;border-top:1px solid var(--border);color