Best possible solution:

Land the focused owner-boundary assertions and remove the duplicate temp-home suite once exact-head CI confirms every distinct contract remains covered.

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

Not applicable: this PR addresses duplicated test execution rather than a runtime defect, and the performance comparison is documented with Testbox runs.

Is this the best way to solve the issue?

Yes. Moving distinct assertions to the model-selection and forwarding owner suites, then deleting the overlapping integration harness, is cleaner than retaining duplicate end-to-end-style coverage.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against cae29a079203.

Label changes:

• add P3: This is a low-risk, test-only CI performance cleanup with no intended runtime or user-visible behavior change.
• add rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🌊 off-meta tidepool and patch quality is 🦞 diamond lobster.
• add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: Real user-behavior proof is not applicable to this maintainer-labeled, behavior-neutral test consolidation; exact-base Testbox timing and focused-suite results are the appropriate evidence.

Label justifications:

• P3: This is a low-risk, test-only CI performance cleanup with no intended runtime or user-visible behavior change.
• rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🌊 off-meta tidepool and patch quality is 🦞 diamond lobster.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: Real user-behavior proof is not applicable to this maintainer-labeled, behavior-neutral test consolidation; exact-base Testbox timing and focused-suite results are the appropriate evidence.

PR surface:

Tests -241. Total -241 across 3 files.

What I checked:

• Focused coverage retained: The branch adds explicit Gmail hook precedence assertions to the model-selection suite and default-thinking forwarding coverage to the run-forwarding suite. (src/cron/isolated-agent.model-formatting.test.ts:524, 28d2f0422d9a)
• Runtime selection contract: Current model selection applies an allowed Gmail hook model before stored session overrides, matching the focused replacement assertions. (src/cron/isolated-agent/model-selection.ts:73, cae29a079203)
• Thinking forwarding contract: The isolated cron runner resolves the configured thinking default for the selected provider/model and carries it into the prepared embedded-run context. (src/cron/isolated-agent/run.ts:331, cae29a079203)
• Validation evidence: The PR reports exact-base Testbox replacement-suite coverage of 58/58 in 5.58 seconds, prior changed checks passing, and a fresh autoreview with no accepted findings. (28d2f0422d9a)
• Current-main necessity: Current main still contains the broad model-overrides integration file, so the branch's test-runtime cleanup has not already been implemented. (src/cron/isolated-agent.model-overrides.test.ts:1, cae29a079203)
• Area provenance: Isolated cron execution has been repeatedly refactored around the same runner boundary; commit a81dc15 is a representative current-main history point authored by vincentkoc. (src/cron/isolated-agent/run.ts:1, a81dc153c61b)

Likely related people:

• tyler6204: Repository contributor guidance assigns cron ownership to Tyler Yust, making this the clearest routing candidate for test-boundary decisions in this subsystem. (role: cron area owner; confidence: high; files: src/cron/isolated-agent/model-selection.ts, src/cron/isolated-agent/run.ts)
• vincentkoc: Authored merged isolated cron runner refactoring, including commit a81dc15 on the central execution path covered by these tests. (role: isolated-run refactor contributor; confidence: high; commits: a81dc153c61b; files: src/cron/isolated-agent/run.ts, src/cron/isolated-agent/run.types.ts)
• steipete: Has prior merged cron service and performance work and is connected to the current test-runtime consolidation beyond merely opening this PR. (role: recent adjacent cron and performance contributor; confidence: medium; commits: 7877182b6f59, e3dd80f9d4f4, b6be42230672; files: src/cron/service/ops.ts, src/cron/service-contract.ts, src/cron/isolated-agent.model-overrides.test.ts)

• 🦀 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.

• 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.