• ClawSweeper keeps one durable marker-backed review comment per issue or PR.
• Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
• A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
• PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
• Maintainers can also comment @clawsweeper review to request a fresh review only.
• Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
• Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
• Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

• 🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
• 🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
• 🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
• 🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
• 🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
• 🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
• 🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

Best possible solution:

Land one canonical implementation that clears only owner-matched stale runtime pins through doctor owner contracts, then close the linked bug after merge.

Do we have a high-confidence way to reproduce the issue?

Yes from source inspection: current main skips agentRuntimeOverride in the plugin route-state scan gate and pinned-runtime repair only deletes agentHarnessId. I did not run a mutating current-main doctor --fix command in this read-only review.

Is this the best way to solve the issue?

Yes, with maintainer acceptance of the owner metadata: the PR extends the existing doctor owner-contract path and tracks exact matched fields instead of adding startup migration or broad deletion logic. The remaining decision is which overlapping branch should be canonical.

Label justifications:

• P2: This is a normal-priority doctor bug fix for stale persisted session routing state with limited blast radius.
• merge-risk: 🚨 session-state: The PR changes doctor --fix behavior that deletes persisted session-level runtime routing fields when classified as stale.
• rating: 🐚 platinum hermit: Current PR rating is 🐚 platinum hermit because proof is 🐚 platinum hermit, patch quality is 🐚 platinum hermit, and Good normal PR readiness: the implementation is focused, proof is sufficient, and the remaining blockers are maintainer ownership and canonical-branch choice rather than contributor fixes.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (live_output): The PR body includes copied live output from a real doctor --fix run against a temporary persisted session store, including before/after JSON showing the stale runtime override removed.
• proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes copied live output from a real doctor --fix run against a temporary persisted session store, including before/after JSON showing the stale runtime override removed.

What I checked:

• Current-main source gap: Current main's plugin session route-state scan gate checks agentHarnessId but not agentRuntimeOverride, and pinned-runtime repair only clears agentHarnessId. (src/commands/doctor-session-state-providers.ts:107, 79be9401306f)
• PR scanner and repair change: The PR includes agentRuntimeOverride in the scan gate, tracks exact pinned runtime keys, and clears only those matched keys during repair. (src/commands/doctor-session-state-providers.ts:117, 4031eff56130)
• PR repair application: The repair loop now iterates pinnedRuntimeKeys, avoiding broad deletion of unrelated runtime override fields. (src/commands/doctor-session-state-providers.ts:413, 4031eff56130)
• Owner metadata wiring: The PR registers Anthropic/Claude CLI and Google/Gemini CLI route-state owners through plugin doctor contracts. (extensions/anthropic/doctor-contract-api.ts:5, 4031eff56130)
• Regression coverage added: The PR tests runtime-override-only stale pins, keeping configured owner runtime pins, clearing stale runtime pins when the provider remains configured, and preserving non-owner runtime overrides. (src/commands/doctor-session-state-providers.test.ts:300, 4031eff56130)
• Related work context: The PR body closes doctor --fix: cleans only codex-route legacy agentRuntimeOverride pins; stale claude-cli/other harness pins persist forever #83098, and the provided context shows overlapping draft fix(doctor): clear stale runtime override pins #83149 targeting the same stale runtime pin cleanup.

Likely related people:

• Moeed Ahmed: git blame in this shallow/grafted checkout attributes the current doctor session route-state provider and Codex doctor contract lines to commit 9c00268914; deeper ownership history is not available locally. (role: current route-state implementation contributor; confidence: low; commits: 9c00268914b9; files: src/commands/doctor-session-state-providers.ts, extensions/codex/doctor-contract-api.ts, src/plugins/doctor-contract-registry.ts)

Codex review notes: model gpt-5.5, reasoning high; reviewed against 79be9401306f.

• Eggs appear after the PR passes real-behavior proof. It is here for vibes, not verdicts: it does not change labels, ratings, merge decisions, or automation.
• The shell reacts to review momentum: open follow-up work warms it up, re-review makes it wobble, and a clean final review lets it hatch.
• Hatchability usually comes from sufficient real-behavior proof, no blocking P0/P1/P2 findings, no security attention needed, and clean correctness. A merged PR is already final, so merge makes the egg hatchable independently.
• The hatch is seeded from this repository and PR number, so the same PR keeps the same creature; the reviewed head SHA can only change safe visual details.
• Rarity is just collectible sparkle: 🥚 common, 🌱 uncommon, 💎 rare, ✨ glimmer, and 🌈 legendary.