Repository navigation
Audit fixes: metric convention required, publish guard, CI contracts job - #298
Merged
Merged
Conversation
…rd, CI contracts job Framework audit (2026-10-18) found three gaps; all fixed TDD-first: - G1: CI never ran isxml validate — new contracts job; structural problems fail, license blocks report BLOCKED and are enforced at publish time (CI stays green while owner decisions are open) - G2: gates now REQUIRE a metric convention (row-mean | corpus-level). The fas five-arm saga (0.60-0.81% vs a recorded 0.52%) proved a gate number without its aggregation is not a gate. All recipes pinned. - G3: publish last-mile guard (pipeline/publish.py) — refuses moving versions, blocked licenses, failing gates, or missing conformance fixtures; the legacy publisher stays the invoked path. Suite 241 green, ruff clean.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
The 2026-10-18 framework audit found three gaps between the WO52 spec
and reality; all are fixed here, tests-first (7 new tests watched
failing before implementation):
isxml validatenow runsas a CI job. Structural violations fail the build; license blocks
print
BLOCKEDand stay non-failing in CI (they're owner decisions,enforced hard at publish time by the G3 guard).
(five reproduction arms at 0.60–0.81% against a recorded 0.52%,
root-caused to row-mean vs corpus-level CER) proved a gate without
its metric convention is not a gate.
GateSpec.conventionis nowrequired; every shipped recipe pins it (fas: corpus-level, others:
row-mean).
pipeline/publish.pyrefuses to publish on: a moving/absent semver (versions are the
owner's decision), any blocked license, any failing gate, or a
missing conformance vectors file. The proven legacy publisher
remains the invoked path; this layer decides whether it may run.
Test plan
tests/test_audit_fixes.py(watched RED first)isxml validateCI-mode smoke:0 problem(s), 3 owner-blocked,exit 0