Version Packages

[khan-actions-bot]

This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will be updated.

Releases

review@1.4.0

Minor Changes

• 4a57d01: Failure scenario required on every finding. Each finding (not just blocking ones) now carries a concrete failure_scenario: the specific inputs or state and the wrong outcome they produce. The field is required in the structured finding schema (FINDING_SCHEMA_VERSION 2), emitted by every producer (label-shape reviewers and all 11 specialist lenses), and carried verbatim into claims.json; the claim-validator attacks exactly that stated scenario, and a scenario too vague to check caps at plausible. Every corpus fixture is migrated with a hand-written failure scenario.

• 11666ee: Reviewer prompt disciplines plus measured cost fixes.

  • Quote-the-rule discipline. A skill or conventions violation may only be flagged when the exact rule text and the exact violating line can both be quoted (skill-auditor, conventions reviewer, and every lens-owned skill application); the validator never confirms a skill claim that cannot quote its rule. No spirit-of-the-doc inference.
  • Amplification nuance. A claim about a mechanism that predates the diff confirms only when the diff materially amplifies its consequence, and the finding must say so explicitly; the correctness reviewer states introduce-vs-amplify on every pre-existing-bug finding.
  • Method-angle procedures. The correctness reviewer works the diff through three named procedures instead of one flat checklist: a line scan, a removed-behavior audit (name the invariant every deleted line enforced; find where the new code re-establishes it), and a cross-file trace of changed symbols' callers and callees.
  • Graceful budget exhaustion. Nearing the run's hard ceilings (per-run AI-credits cap, job timeout), tracked via observable proxies since the agent cannot see its own credit spend (elapsed wall-clock, dispatch counts, the investigation journal), the orchestrator sheds remaining work in a fixed order (opt-in reviewers/lenses first, then the risks/patterns comment, then validation) and submits the verdict from findings validated so far plus skipped-dimension notes, instead of dying at a ceiling with everything spent and nothing posted.
  • Batched safe-output tail. The orchestrator emits same-kind safe outputs together in as few calls and turns as possible; thread resolutions especially, which previously went out one turn apiece.

• 0892c8d: Change-provenance gate, enforced in code, plus generated-stripped whole-change diffs. New lib/diff.ts (unified-diff parsing) and lib/provenance.ts (gate + CLI) compute a per-file changed-line map (provenance.json); a finding whose anchor is not an added or modified line of the diff cannot carry a blocking label and does not post to the PR at all; the set-aside pre-existing observations are recorded in the run artifact (out/pre-existing.json) only. Deletion findings pass via removedAdjacent lines. The gate fails open (gates nothing, with a review-body note) whenever the parsed map cannot be trusted: an unparseable diff, hunks not attributable to a file section, or a changed file whose patch is missing from the parse (files.json now carries hasPatch for this cross-check). The provenance CLI also stages full-stripped.diff (the full diff minus files the router classifies linguist-generated; routing.json now exposes generatedFiles), and every whole-change reviewer and specialist lens reads it instead of the full diff; pattern-triage keeps the full diff since classification is its job. Wired through the no-post eval runner and covered by unit tests plus a smoke corpus case.

• 2cdee74: Thumbs-sweep: opt-in reaction seeding and self-reaction filtering. With seedReactions: true in the sweep config, each sweep seeds one thumbs-up and one thumbs-down (from the bot) on any reviewer comment lacking them, so readers see the feedback affordance without opening the reaction picker. Feedback counts now exclude the bot's own reactions everywhere, so seeded nudges never trigger follow-ups or count as signal; a reaction with no attributed login still counts as a real user's. The port gains an optional addReactions method, required only when seeding is enabled (enabling it without an implementation fails loudly).

Patch Changes

• 8b68f13: Expand the allowed-domains allowlist for the review workflow's safe outputs so links we routinely use in PRs survive gh-aw's text-sanitization. Chosen from the domains that actually appear in our PR bodies and comments (surveyed across recent Khan/frontend PRs): GitHub, khanacademy.org (incl. per-PR deploy previews), khanacademy.dev, Jira/Confluence, Slack, Claude (claude.ai + claude.com), Figma, Google Docs, and Cursor. Entries are bare hosts, which match the host and all of its subdomains.

check-for-changeset@2.0.1

Patch Changes

• Updated dependencies [a28b537]
  • get-changed-files@3.0.1

gerald-pr@6.1.1

Patch Changes

• Updated dependencies [a28b537]
  • get-changed-files@3.0.1

get-changed-files@3.0.1

Patch Changes

• a28b537: Fix a crash (core.warn is not a function) on push events when the pushed commit is associated with more than one open PR, as happens with stacked PRs. The github-script core API method is core.warning, not core.warn.