feat: Stage 2 Phase 3 — labelled corpus seed + executable cull (offline prove) - #3
Merged
Merged
Conversation
Phase 3 advances Stage 2 precision work without more harness packaging: - corpus/seed.jsonl: 22 clean + 15 defective real hunks (requests, httpx, click, urllib3) meeting the pre-registered shape floor - decide_cull / fail_on_gate encode the 2026-08-26 thresholds so the cull is a measurement against fixed numbers, not taste - corpus_cli.py prove: offline proof (shape + 6/6 cull self-check); live run stays blocked without ANTHROPIC_API_KEY - Action fail-on remains never — metrics do not justify flipping it
lazycoder review — 💥 ERROR0 finding(s), 0 abstention(s), 0 rule error(s) — R1..R17 rubric via lazycoder |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Stage 2 (corpus / precision) offline half — not more harness packaging.
corpus/seed.jsonl: 22 clean / 15 defective from merged OSS PRs (requests, httpx, click, urllib3) — meets the pre-registered shape floor (≥20 / ≥15).decide_cull+fail_on_gateencode the 2026-08-26 thresholds fromdocs/hardening-plan.md.python scripts/corpus_cli.py prove corpus/seed.jsonl→ exit 0, artifactdocs/eval-runs/corpus-prove.txt.corpus_cli.py runis BLOCKED withoutANTHROPIC_API_KEY(exit 2). No invented precision numbers.fail-onstaysnever— metrics do not justify flipping it.Proof
uv run python scripts/corpus_cli.py prove corpus/seed.jsonl # EXIT 0 — OFFLINE_PROVE: PASS LIVE_SCORING: BLOCKED FAIL_ON: neverTest plan
pytest -qgreen locally (114 passed)