Repository navigation
The gap sweep caught a false claim in the brief that argued against them - #40
Merged
Merged
Conversation
StepenkoAnatoli
commented
Sep 21, 2026
StepenkoAnatoli
commented
Owner
Sweeping for remaining gaps found one I had committed an hour earlier. The sea-assets brief said it fails `--strict` "because the delivery-architecture corpus passes `--strict` and a reader would otherwise assume parity". Nothing in this repository passes `--strict`. Measured on main: root PASS 7 warn -> FAIL 7 blocking 2026-09-21-delivery-architecture PASS 8 warn -> FAIL 8 blocking 2026-09-21-agent-interface PASS 5 warn -> FAIL 5 blocking 2026-09-21-public-run-visibility PASS 4 warn -> FAIL 4 blocking 2026-09-21-sea-assets PASS 2 warn -> FAIL 2 blocking The delivery corpus did pass `--strict` - before ADR-0036 added `corroboration` to POLICY_CHECKS. The sentence was written from that memory instead of from a run, inside a brief whose subject is a claim asserted without a source. Caught by sweeping, not by any check. ADR-0036 verified "every existing corpus stays green under the default policy" and stopped there. That was the wrong half: the change was to what `strict` means, and nothing measured what it now costs. All seven root warnings are `corroboration` findings, so `evidencePolicy=strict` fails identically - confirmed by running it with a config rather than inferred: `FAIL 7 blocking, 0 warning(s), 18 passing [evidencePolicy=strict]`. Recorded, not fixed, deliberately. The check is right on every finding; the corpora genuinely are single-sourced. Making `strict` passable means either collecting second sources for thirty-odd unknowns or softening a check that is telling the truth, and neither belongs in an amendment about measurement. What an operator needs is the honest label: `strict` is aspirational here, not a standard this repository meets, and its findings are a worklist not a regression. Also separated two settings the verdict line conflates. `--strict` promotes every warning; `evidencePolicy=strict` promotes only the three POLICY_CHECKS. The summary prints the declared policy either way, so a `--strict` run reads `FAIL ... [evidencePolicy=pluralist]`, which looks like pluralist failed it. ADR-0035's "What is NOT measured" section gains a superseded marker. It was closed by the amendment directly below it, but a top-down reader stopped at "**unknown**". 57/57 tests pass. All four corpora PASS and every chain verifies. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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.