From 05fcaac755adb334340bbc72a14c7a71ca930a82 Mon Sep 17 00:00:00 2001 From: Jeff Smith Date: Mon, 6 Apr 2026 21:47:49 -0600 Subject: [PATCH] docs(plan): note classifier rebuild (#42) in Phase 2 The filetype classifier is biased toward source code and would mislead the survey pass on non-code targets (mail, notebooks, ledgers). #5 ships with a prompt-level Band-Aid; #42 captures the real fix and is sequenced after the survey pass is observable end-to-end and before Phase 3 depends on survey output. --- PLAN.md | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/PLAN.md b/PLAN.md index fdab07b..a74c83c 100644 --- a/PLAN.md +++ b/PLAN.md @@ -535,6 +535,11 @@ without 9 phases of rework. - `_SURVEY_SYSTEM_PROMPT` in `prompts.py` - Wire into `_run_investigation()` before dir loops - Survey output injected into dir loop system prompt +- **Rebuild filetype classifier (#42)** to remove source-code bias — + lands after the survey pass is observable end-to-end (#4–#7) and + before Phase 3 starts depending on survey output for real decisions. + Until then, the survey prompt carries a Band-Aid warning that the + histogram is biased toward source code. ### Phase 3 — Investigation planning - Planning pass after survey, before dir loops