feat(qa): L3 answer-packet state schemas, surface-contract spec, first evidence folder#576
Merged
Conversation
…spec The contract gains "verifiable meaning": an answerPacket section declares, per response shape x evidence support, which packet fields MUST appear, which must NEVER leak (compact rendering is driven by field emptiness, so a leak is a rendered regression), and which honesty markers are required (Source needed prefixes, limitation risk rows). A new executor drives every declaration through the REAL detectRequestedResponseShape + applyDeterministicResponsePolicy — 8 shapes x supported/unsupported plus detector agreement, 25 tests from one JSON file — and joins the runtime-smoke allowlist. SPEC.md distills the pattern to one portable page (three layers, schema, verifier contract incl. independent replay, fail-closed hash binding for served projections) so the second in-portfolio implementation can converge on one spec instead of two bespoke ones. Reversion-proved: flipping the bullets lineCount clause fails exactly the two bullets tests; restored, 25/25. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…after Five before/after pairs (desktop+mobile x dark+light + the forced context-miss state) regenerated honestly: before = pre-#574 sources rebuilt and captured with the FIXED spec so both sides render true dark. Schema-valid manifest per the ui-contract folder convention. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
HomenShum
enabled auto-merge (squash)
July 17, 2026 21:02
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
HomenShum
changed the title
HomenShum
changed the title
|
Demo: walkthrough of the surfaces this PR changed is available as a workflow artifact ( |
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
2 participants
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 (corrected post-merge: this PR merged with exactly two commits)
answerPacketsection declares, per response shape × evidence support, which packet fields MUST appear, which must NEVER leak, and which honesty markers are required.chatRuns.contract.test.tsexecutes every clause through the real runtime functions — 25 tests from one JSON file — and joins the runtime-smoke allowlist. Reversion-proved.20260717-design-nitpick/with five schema-valid before/after pairs regenerated from pre-fix(redesign): design nitpick pass — honest dark captures, state copy, icons, row layout #574 sources.Follow-up coverage/judge/nightly work ships separately in the next PR (its content was briefly described here by mistake while that commit was unmerged).
🤖 Generated with Claude Code