ci: route control-plane lanes through the governed selector - #237
Conversation
Route ccp's remaining GitHub-hosted control-plane lanes through the governed runner selector to the self-hosted fleet, ending hosted-minute spend on this private repo: - ci.yml ci-status and pr-title.yml pr-title: the fail-closed required gateways route via the selector instead of a fixed ubuntu-24.04 runner (fail-closed on selector/fleet outage is the accepted tradeoff). - pr-title validate-pr-title: bump semantic-pr to its runner-input variant, passing runner + prerequisite-result. - link-check: migrate to the runner-input link-check reusable behind a select-runner job; retire the strict-policy pause. - Add do-not-merge and pr-issue-linkage callers (ci-workflows#120/#121), both routed through the selector. - Drop all .github/runner-policy.json hosted exceptions (now zero). Refs melodic-software/github-iac#78. 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01K3QehVwmWzkBLpKokNCkkt Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 4621267c29
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
|
Claude finished @kyle-sexton's task in 3m 57s —— View job PR Review — ci: route control-plane lanes through the governed selector
SummaryThe fleet-routing migration itself is structurally sound — fail-closed semantics are applied consistently and the design decisions are well-documented. However, there is one blocking gap that will cause the runner-policy CI lane to fail: the two new reusable workflow SHAs called from 🔴 Blocking — Missing runner-policy contractsCodex's P1 findings are confirmed.
Neither SHA appears in The PR description states these contracts "landed via the standards sync (#236) already merged to main," but #236 does not appear in the branch's git ancestry — the most recent standards sync visible is Both callers follow the Required entries follow the same shape as the existing "melodic-software/ci-workflows/.github/workflows/do-not-merge-gate.yml@885302176345486ca6c2c392d83131f9b5389251": {
"routing": "runner-input",
"runnerInput": "runner",
"selectorResultInput": "prerequisite-result",
"allowedInputs": ["runner", "prerequisite-result"],
"allowedSecrets": {},
"allowedCallerPermissions": {
"pull-requests": "read"
}
},
"melodic-software/ci-workflows/.github/workflows/pr-issue-linkage.yml@f7e94a80254fdca0aa85a600a07d784753b090e3": {
"routing": "runner-input",
"runnerInput": "runner",
"selectorResultInput": "prerequisite-result",
"allowedInputs": ["runner", "prerequisite-result"],
"allowedSecrets": {}
}(exact Resolution: Either merge standards sync #236 first (if it truly contains these contracts), or add them to ✅ What looks correct
Adding
The shift from
The
The security justification in the comments is sound:
Dropping all 3 exceptions ( 🟡 Minor — select-runner SHA version skewNew workflows ( Bottom lineMerge is blocked on the missing |
…rine PLUGIN-PHILOSOPHY's design boundary is the plugin specialization of the org-wide convention now owned by melodic-software/standards conventions/engineering/shareable-artifact-design.md (standards #237, PR #238). Design boundary states the specialization relationship; authoritative references gain the explicit entry. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…rine (#941) ## Summary PLUGIN-PHILOSOPHY's design boundary is the plugin specialization of the org-wide convention now owned by `melodic-software/standards` `conventions/engineering/shareable-artifact-design.md` (standards #237, merged in standards PR #238). The design boundary states the specialization relationship; the authoritative references gain the explicit entry. Completes the follow-up promised in #933's companion work. No linked issue: the tracked items (#933, standards#237) are already closed by their own PRs; this is their promised follow-up citation. ## Related - melodic-software/standards#237 — the convention doc this cites (closed by standards PR #238; not closed by this PR). - #933 — the advisor-escalation work this citation follow-up was sequenced after (already closed by #934). 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
…#1552) ## Summary This repository is where both silent-review blackouts were measured, and it is the one repository still blind to them. It pinned the review reusable at `90f1c54` (2026-07-18) and the security-review reusable at `99cb082` (2026-07-21) — both predate the infra-failure classifier. When the review lane fails here it reports green and says nothing about why. Bumps both to `ci-workflows@e295107`, which carries: - **melodic-software/ci-workflows#248** — infra failures classify as `auth` | `rate-limit` | `overloaded` | `other`, emitted as a bare `class=<token>` term in the job's `::error` annotation and a `Failure class:` line in the marker-managed infra-status PR comment, with the numeric `api_error_status` alongside in the safe structured projection. - **melodic-software/ci-workflows#249** — the substring allowlist mirrors the status mapping type-for-type, so a payload with no recoverable numeric status lands in the class the status would have chosen. **Gate behavior is unchanged.** Passing the check on an infrastructure failure remains an explicit non-goal upstream; this bump adds diagnosis, not blocking. Nothing about which checks gate a merge here changes. ### Why it matters here specifically The measured impact on this repo, from the investigation on ci-workflows#228: - 2026-07-25T22:48:55Z → 2026-07-26T01:40:48Z, ~2h52m - **158 of the 159** review runs launched in that window failed — a complete census, every run's step conclusions inspected — and every one reported job conclusion `success` - **36 of the 38** PRs merged in that window merged with a silently failed review - A second blackout was still live at the time of writing, with every run since 09:02Z failing Throughout all of it, the only signal was a warning comment. There was no machine-readable cause, and no way to tell a revoked credential from a rate limit without guessing from `duration_ms`. ### Interface compatibility — checked, not assumed - `claude-review.yml`: the `workflow_call` block is **byte-identical** across `90f1c54..e295107`. - `claude-security-review.yml`: differs only in the prose of an input *default* — the security prompt now defers to zizmor's static lane for supply-chain, dangerous-trigger, excessive-permission, and template-injection findings, so this lane focuses on logic, architecture, data-flow, and trust-boundary reasoning. No input or secret was added, removed, or renamed in either. ## Test plan - Interface diff of both reusables' `workflow_call` blocks across the pin jump — verified above, no breaking change. - Confirmed the target SHA exists and that both reusables at `e295107` contain the class emission (`gh api .../contents/...?ref=e295107`). - Upstream verification at the source: 273 node tests and 23 shell cases passing on `ci-workflows@e295107`, with the classification branches and the no-leak canary mutation-verified. - Already proven end-to-end in production: the lane infra-failed on ci-workflows#248's own run and emitted `class=rate-limit` with `"api_error_status":429`, on the exact `subtype: success` / `is_error: true` shape both blackouts produced. - **This PR's own review runs are the live check.** If the lane is still rate-limited when CI runs, the infra-status comment on this PR should now name the class instead of leaving it unexplained — which is the whole point of the bump. ## Related No linked issue. Upstream work is melodic-software/ci-workflows#228, #237, #248, #249. The prior diagnosis in this repo, #1122, is closed; this is the consumer-side adoption of the fix that came out of it, not new work needing its own tracking item. Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Wave 3 per-repo floor conversion (melodic-software/github-iac#78, 2026-07-16 owner override): converts every remaining hosted control-plane lane to governed selector routing — claude-code-plugins ends with ZERO runner-policy exceptions.
No linked issue.
What
ci.yml: ci-status aggregate routes withif: ${{ !cancelled() }}(fails closed through execution on the hosted fallback, never by skipping); the workflow-schemafiles:list gains the two new callers.pr-title.yml: the required-check status wrapper routes to the fleet; semantic-pr pin bumped to51012e2withprerequisite-result(if: ${{ always() }}per the validator's fail-closed reporter contract).link-check.yml: select-runner added; pin bumped to the3dfb184runner-input variant; retired theCI_RUNNER_POLICY != self-hosted-onlypause guard.do-not-merge+pr-issue-linkagecallers (Roll out do-not-merge-gate caller to all requires-ci repos ci-workflows#120 / feat(work-items): republish v2 plugin from post-seam provider-neutral core #121 rollout):pull_request_target+merge_group; execute on the hosted fallback until select-runner: admit pull_request_target and merge_group metadata-only events for fleet routing ci-workflows#130 admits those events.runner-policy.json: all 3 exceptions dropped →"exceptions": {}.Verification
runner-policy.mjsrun (CI env parity): clean; the two new callers' contracts landed via the standards sync (chore: sync standards components #236) already merged to main.runs-onkeeps the|| 'ubuntu-24.04'fallback (epic decision 4).Related
🤖 Generated with Claude Code
https://claude.ai/code/session_01K3QehVwmWzkBLpKokNCkkt