Narrow the blocking gate to hard facts and add fact-consistency checks (quality P5)

Conversation 0 Commits 2 Files Changed 22

+416 -48

2 Commits

Author	SHA1	Message	Date
weichen	c802be812d	Merge origin/main into feature/agent-quality-phase05 ci / vet-and-test (pull_request) Successful in 40s Details	2026-10-10 00:30:05 +00:00
weichen	ef060d4ee5	Keep only hard facts in the blocking gate and add fact-consistency checks ci / vet-and-test (pull_request) Successful in 40s Details Why: User rulings 4/5 plus the P2-P4 live evidence. Keyword scoring dimensions (readability, actionability) kept failing valid documents - normal Chinese section headings flagged as column titles, loops of 4 corrections - because they grade writing preferences, not verifiable facts. Unverified numbers (~10-13 per run) remained the largest quality leak through every phase. And the verifier runs on the same model as the worker (lightweight falls back to chat), so "independent" verification was self-review. Separately, svc seeding after boot never registered the tavily tools until a restart because keys were snapshotted at startup. Changes: quality.HardFacts (lint actually ran + zero errors) replaces the full ScoreBytes projection in CheckArtifacts - keyword dimensions remain in ScoreBytes for eval visibility only. New fact-consistency check (runtime.factCheck, default on): extract percentages/dates/large numbers from deliverable text, trace each against the material corpus (attachment text + tool evidence); misses get exactly one correction round (fix, drop, or mark as target/estimate), never a loop. runVerifier skips same-source model verification unless runtime.verifierSameModel is set (hard facts + fact check still run). Search tools now register lazily, reading tavily keys per call (settings/seeding take effect immediately). Evidence: SELF - TestHardFactsIgnoresKeywordDimensions, TestHardFactsKeepsLintUnavailableBlocking, TestUnverifiedFactsAndNotice, TestFactCheckContinuationOnce (fact note fires, corrected run passes); go vet + go test ./... pass; live before/after on the golden task set recorded on the phase05 task card per the policy clause. Impact: Gate stops rejecting valid documents on stylistic grounds; invented figures get one forced correction round; same-model verification no longer pretends to be independent; tavily config is hot. No contract or storage changes.	2026-10-10 00:04:57 +00:00

Author

SHA1

Message

Date

weichen

c802be812d

Merge origin/main into feature/agent-quality-phase05

ci / vet-and-test (pull_request) Successful in 40s

Details

2026-10-10 00:30:05 +00:00

weichen

ef060d4ee5

Keep only hard facts in the blocking gate and add fact-consistency checks

ci / vet-and-test (pull_request) Successful in 40s

Details

Why: User rulings 4/5 plus the P2-P4 live evidence. Keyword scoring dimensions
(readability, actionability) kept failing valid documents - normal Chinese
section headings flagged as column titles, loops of 4 corrections - because
they grade writing preferences, not verifiable facts. Unverified numbers
(~10-13 per run) remained the largest quality leak through every phase.
And the verifier runs on the same model as the worker (lightweight falls back
to chat), so "independent" verification was self-review. Separately, svc
seeding after boot never registered the tavily tools until a restart because
keys were snapshotted at startup.

Changes: quality.HardFacts (lint actually ran + zero errors) replaces the
full ScoreBytes projection in CheckArtifacts - keyword dimensions remain in
ScoreBytes for eval visibility only. New fact-consistency check
(runtime.factCheck, default on): extract percentages/dates/large numbers from
deliverable text, trace each against the material corpus (attachment text +
tool evidence); misses get exactly one correction round (fix, drop, or mark
as target/estimate), never a loop. runVerifier skips same-source model
verification unless runtime.verifierSameModel is set (hard facts + fact check
still run). Search tools now register lazily, reading tavily keys per call
(settings/seeding take effect immediately).

Evidence: SELF - TestHardFactsIgnoresKeywordDimensions,
TestHardFactsKeepsLintUnavailableBlocking, TestUnverifiedFactsAndNotice,
TestFactCheckContinuationOnce (fact note fires, corrected run passes);
go vet + go test ./... pass; live before/after on the golden task set
recorded on the phase05 task card per the policy clause.

Impact: Gate stops rejecting valid documents on stylistic grounds; invented
figures get one forced correction round; same-model verification no longer
pretends to be independent; tavily config is hot. No contract or storage
changes.

2026-10-10 00:04:57 +00:00

Narrow the blocking gate to hard facts and add fact-consistency checks (quality P5) #123