Best possible solution:

Have maintainers approve the runtime identity contract and land this file-backed seam before dependent SQLite, SDK, and reader slices build on it.

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

Not applicable; this is a staged refactor contract rather than a bug report with a failing reproduction. The useful checks are source review, focused file-backed tests, and maintainer review of the contract semantics.

Is this the best way to solve the issue?

Yes, if maintainers accept the linked Path 3 plan: a file-backed identity seam is narrower than combining caller migration with the SQLite storage flip. It still needs human contract approval because downstream slices rely on this shape.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against 8ded75628437.

Label changes:

• remove proof: sufficient: Current real behavior proof status is not_applicable, not sufficient.

Label justifications:

• P2: This is normal-priority internal migration work with limited immediate user-facing blast radius but meaningful session-state implications.
• merge-risk: 🚨 session-state: The diff changes transcript/session identity and mutation helpers that could mis-associate runtime transcript state if the contract is wrong.
• rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🌊 off-meta tidepool and patch quality is 🐚 platinum hermit.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Not applicable: The maintainer-labeled internal architecture slice is not subject to the external contributor real-behavior proof gate; focused tests and maintainer contract review are the relevant proof path.

PR surface:

Source +371, Tests +266, Other +6. Total +643 across 14 files.

What I checked:

• Repository policy applied: Root and scoped AGENTS.md were read; session state, migration seams, gateway helpers, scripts, and tests are compatibility-sensitive review areas for this PR. (AGENTS.md:24, 8ded75628437)
• Current main does not already implement the contract: Current main has no runtime transcript scope/target module or helper symbols from this PR; the only transcript-runtime match is unrelated runtime plan policy typing. (8ded75628437)
• PR head adds runtime transcript identity: The PR head adds SessionTranscriptRuntimeScope/Target plus separate write and read target resolution, keeping sessionFile as a file-backed implementation detail. (src/config/sessions/session-accessor.ts:88, 7ffe7061dccb)
• PR head adds runtime transcript state facade: The new facade exposes target resolution, read, persist, replace, existence, and delete helpers over the resolved runtime transcript target. (src/agents/embedded-agent-runner/transcript-runtime-state.ts:30, 7ffe7061dccb)
• PR head adds scoped mutation wrappers: The PR wraps transcript rewrite, tool-result truncation, successor rotation, and checkpoint snapshot reads around runtime transcript scope while leaving storage file-backed. (src/agents/embedded-agent-runner/transcript-rewrite.ts:383, 7ffe7061dccb)
• Focused tests cover the key identity invariant: PR-head tests cover topic-session target alignment and missing-transcript probes that must not create session metadata. (src/config/sessions/session-accessor.test.ts:588, 7ffe7061dccb)

Likely related people:

• jalehman: Live PR and related tracker history show this person carrying the Path 3 session/transcript identity stack; merged PR refactor: add session accessor seam with gateway consumer #90463 introduced the current session accessor seam that this PR extends. (role: feature stack owner and recent area contributor; confidence: high; commits: 7ffe7061dccb, fee584a07404, 58aa59eaf805; files: src/config/sessions/session-accessor.ts, src/agents/embedded-agent-runner/transcript-runtime-state.ts, src/gateway/session-compaction-checkpoints.ts)
• osolmaz: The linked tracker discussion connects this migration to false transcript-takeover and message-loss clusters, providing maintainer context for why the session/transcript identity boundary matters. (role: adjacent behavior reporter; confidence: medium; files: src/agents/embedded-agent-runner/run/attempt.session-lock.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.