Commit Graph
3 Commits
Author SHA1 Message Date
d82df27473 Phase 3: canvas smoke-test evidence and weekly review metrics (#4184) (#4191)
* Phase 3: canvas smoke-test evidence, safe auto-merge, and weekly review metrics (#4184)

Adds the canvas-smoke-test check with review evidence, a config-gated safe auto-merge workflow (disabled by default), and a weekly review operating metrics workflow, plus maintainer docs.

Refs github/awesome-copilot#4184

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Trigger safe auto-merge sweep after Submission Gate Writer completes

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Rename thirdParty identifier to satisfy codespell

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Address Copilot review on canvas smoke test, auto-merge, and metrics

- Bind canvas report writer to the triggering workflow_run PR (base/head repo, ref, SHA)
- Bound PNG inflation by IHDR-derived size and a decoded-bytes budget
- Classify file: URLs as unsafe manifest paths
- Follow CommonJS require() and literal dynamic imports; error on missing
  reachable targets and devDependency-only runtime imports
- Validate removed extension paths and orphaned plugins instead of skipping
- Warn on .py/.rb/.pl scripts
- Auto-merge: support CODEOWNERS teams (fail closed), front matter authors,
  list PRs by head SHA, only fall back to direct merge on clean status,
  drop ineffective check_run trigger
- Metrics: ignore reviews before the PR was ready for review

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Docs: note microsoft/azure-dev-tools as a portable marketplace for the review model

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Auto-merge: only trust submission-gate runs from the Submission Gate Writer

Adds trusted_checks (default submission-gate -> external_id
submission-gate-writer). Any other check run or status with that name on the
head SHA blocks arming, so a PR-defined job cannot satisfy the gate.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Harden canvas smoke-test checker

- String-, template-, and regex-aware comment stripping so literals containing // cannot hide imports
- Never follow symlinks or read non-regular files; bounded, contained reads
- Validate plugins whose plugin.json was deleted
- Reject unsafe extension IDs in plugin manifests before materialization

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Phase 3: remove auto-merge, isolate canvas smoke test, harden checker

- Remove safe auto-merge (config, workflow, script, tests, labels) per
  maintainer review; documented as deferred pending a security review.
- Restore validate-canvas-extensions.yml to main and move the smoke test
  to its own canvas-smoke-test.yml workflow scoped to extensions/**.
- Reject duplicate IHDR chunks in preview PNGs.
- Mask string/template/regex literal contents before extracting dynamic
  import() and require() specifiers.
- Resolve package.json imports aliases through the same classifier and
  module graph; fail closed on alias shapes that cannot be analyzed.
- Rename docs to canvas-evidence-and-metrics.md and update references.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fix canvas smoke and review metrics findings

- Trigger canvas smoke tests for plugin changes while skipping expensive setup when target detection finds no canvas coverage.
- Fail reachable data: imports while preserving warnings for unreachable module data: imports.
- Use base plugin manifests for deleted bundle plugin detection and fail closed when base manifests cannot be read.
- Paginate GraphQL PR review pages before computing weekly review metrics.
- Report workflow run API collection errors as incomplete automation health instead of not found.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Address follow-up canvas review findings

Fail closed on non-literal runtime imports, require canonical PNG termination, restrict comment updates to github-actions[bot], and paginate external-plugin label events.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Aaron Powell <me@aaron-powell.com>
2026-10-07 16:36:13 +11:00
James MontemagnoandCopilot App d6131471b8 Phase 2: submission-gate, risk tiers, and PR status state machine (#4184) (#4190)
* Add submission-gate check, merge-risk tiers, and PR status state machine (#4184)

Phase 2 (enforcement) of github/awesome-copilot#4184:
- submission-gate aggregate required check (reader) with infra vs contribution failure classification
- merge-risk:low/medium/high classification from .github/risk-tiers.yml with tier approval policy
- Submission Gate Writer maintaining state labels and a persistent status comment
- /rerun-checks and /request-review PR commands
- validate-skills and validate-submission-gate workflows, unit tests, and maintainer docs

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fix codespell findings and add spelling to submission gate checks

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Harden submission gate per review feedback

- Publish the submission-gate check from the trusted writer via the Checks API
  and flag any other submission-gate check run as tampering
- Fail closed on truncated file lists, unscannable diffs, permission lookup
  errors, and skipped required checks
- Discard evaluations when the PR head or reviews change mid-run
- Low tier requires a resource owner approval
- Commands must start the comment, matching the workflow filter

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Address #4190 review: command reader/writer, pools, paths, staleness

- Split /rerun-checks and /request-review into a read-only issue_comment
  reader and a workflow_run writer that re-reads the comment, PR, and
  permission before writing.
- Drop the workflow-security pool; agentic workflows and hooks are
  content (medium unless they add scripts or hook commands); the core
  pool is core-maintainers.
- Include each check's workflow file and eng scripts in its paths.
- Discard evaluations when the base branch or risk labels change; run
  the gate on PR edits (retargeting).
- Correct copies of the writer's check that claim success while the
  evaluation fails.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fail closed on lost contributor signal; publish gate check before labels

- A contributor check whose pr-check job succeeded but whose result artifact
  is missing or unreadable is now an infrastructure failure, so a lost HIGH
  signal can't lower the approval tier.
- Publish the submission-gate check right after stale-state revalidation;
  label and comment sync are independent and reported afterwards.
- Revert the self-path added to validate-agentic-workflows-pr.yml's trigger:
  that workflow rejects any .github change. Exempt it in the drift test.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fail closed on bad routing config; advisory checks never hold the gate

- loadGateConfig: a missing review-routing.yml still falls back, but a
  file that can't be parsed (or whose pools isn't a mapping) now throws
  instead of silently relaxing approvals.
- Pending required: false checks no longer keep the PR in
  awaiting-automation or the gate check in progress, and the polling loop
  doesn't wait on them. They render as pending (advisory).
- PR commands: 👀 marks a command as claimed; 🚀 is written only after
  the reply is posted and is the only replay guard, so a run that failed
  partway can be retried.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Fix codespell: unparseable -> unparsable

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
2026-10-01 11:39:09 +10:00
0c54bcee82 Phase 1: distributed CODEOWNERS and automated reviewer routing (#4184) (#4189)
* Phase 1: distributed CODEOWNERS and automated reviewer routing (#4184)

- Replace the individual default CODEOWNERS owner with the core-maintainer
  team and add domain team ownership for extensions, plugins, content,
  hooks, workflows, .github, and eng.
- Add .github/review-routing.yml as the single source of truth for reviewer
  pools, routes, labels, and SLA settings (ships in dry-run mode).
- Add eng/review-routing.mjs (+ unit tests) with business-day math,
  least-loaded reviewer selection, routing and escalation planning, and
  workflow_run artifact validation.
- Add Review Routing (workflow_run/dispatch/schedule), Review Routing Request
  (read-only needs-reviewer reader), and Review Escalation (weekday schedule)
  workflows.
- Add needs-reviewer, review-overdue, review-escalated to setup-labels.
- Document the model in docs/maintainers/review-routing.md.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Address review feedback on reviewer routing and escalation

- Escalate at routedOn + escalation business days (Mon -> Fri)
- Leave labels untouched when requestReviewers fails and fail the job
- Serialize read/plan/apply per PR across routing and escalation
- Treat Label PR Intent artifact labels as authoritative
- Re-route to an uncovered pool on head changes without restarting SLA
- Use the due-label timestamp as the review boundary
- Fall back to pool/escalation teams and report what was requested

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Address second-round review feedback on review routing (#4184)

- CODEOWNERS: route /hooks/ and /workflows/ to content reviewers, drop the
  separate /.github/workflows/ rule, and leave per-resource owner lines as on
  main (their removal is deferred to a separate cleanup).
- Routing: ignore pending requests for unavailable reviewers when checking
  coverage; needs-reviewer excludes prior reviewers and changes nothing when no
  new reviewer can be requested.
- Escalation comments show the actual routing date.
- Cap sweep matrices at 256 targets and defer the remainder.
- Docs: best-effort load balancing, matrix cap, and needs-reviewer semantics.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Route workflow and hooks labels to the content pool (#4184)

Drop the workflow-security reviewer pool per maintainer feedback: agentic
workflows and hooks are reviewed like other content, and the repository's
own GitHub Actions workflows stay with core maintainers.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Pin routing contract in tests; make CONTRIBUTING routing note dry-run aware

- Assert the exact label-to-pool mapping and priority order, pool set, and default/escalation pools for the shipped config.
- CONTRIBUTING.md: describe automatic routing as active only once dry_run is turned off.

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>

* Simplify dry_run validation condition

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Aaron Powell <me@aaron-powell.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
2026-10-01 00:51:10 +00:00