Best possible solution:

Land after maintainer review and exact-head CI completion, keeping the current source-compatible accessor shape and leaving the broader SQLite backend flip tracked by #88838.

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

Not applicable; this is a behavior-preserving refactor rather than a reported current-main bug reproduction.

Is this the best way to solve the issue?

Yes; the current head uses targeted session accessors for bundled metadata reads while preserving list-session source compatibility and parent-session model override behavior.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against eac1d3349c29.

Label changes:

• add rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🦞 diamond lobster and patch quality is 🦞 diamond lobster.
• add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (live_output): The PR includes live Gateway proof for the changed runtime behavior, plus focused accessor/boundary proof after the compatibility fix.
• remove rating: 🦐 gold shrimp: Current PR rating is rating: 🦞 diamond lobster, so this older rating label is no longer current.
• remove merge-risk: 🚨 compatibility: Current PR review selected no merge-risk labels.
• remove status: ⏳ waiting on author: Current PR status label is status: 👀 ready for maintainer look.

Label justifications:

• P2: This is a bounded cleanup in session metadata/runtime plumbing with limited blast radius, but it touches compatibility-sensitive accessor and bundled plugin paths.
• rating: 🦞 diamond lobster: Overall readiness is 🦞 diamond lobster; proof is 🦞 diamond lobster and patch quality is 🦞 diamond lobster.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (live_output): The PR includes live Gateway proof for the changed runtime behavior, plus focused accessor/boundary proof after the compatibility fix.
• proof: sufficient: Contributor real behavior proof is sufficient. The PR includes live Gateway proof for the changed runtime behavior, plus focused accessor/boundary proof after the compatibility fix.

PR surface:

Source -36, Tests +45, Other +8. Total +17 across 22 files.

What I checked:

• PR state: this PR is open at head 5c0e33b, targets main, and carries the protected maintainer review label; it is not a close candidate while the maintainer-controlled Path 3 tracker still names it as active work. (5c0e33b2b9b7)
• Source-compatible SDK list params: The current head keeps SessionStoreListParams derived from SessionStoreReadParams, so callers that pass readConsistency to listSessionEntries remain source-compatible even though list reads do not use that option. (src/plugin-sdk/session-store-runtime.ts:27, 5c0e33b2b9b7)
• Runtime latest-read forwarding: Runtime session access params now include readConsistency and only forward it through the single-entry getSessionEntry path, preserving the public list parameter shape without changing list semantics. (src/plugins/runtime/runtime-agent.ts:92, 5c0e33b2b9b7)
• Accessor cache contract: The session accessor maps readConsistency: latest to skipCache for single-entry reads, while list entries still use the existing listing path. (src/config/sessions/session-accessor.ts:908, 5c0e33b2b9b7)
• Parent model override preserved: resolveStoredModelOverride now loads parent-session metadata through an injected targeted loader, and the Mattermost caller passes getSessionEntry with latest-read consistency, preserving explicit parent-session inheritance without whole-store access. (src/auto-reply/reply/stored-model-override.ts:35, 5c0e33b2b9b7)
• Boundary ratchet extended: The accessor-boundary guard adds the migrated bundled metadata readers to the allowlisted accessor-only surface, so future whole-store regressions are checked by the existing script and tests. (scripts/check-session-accessor-boundary.mjs:127, 5c0e33b2b9b7)

Likely related people:

• jalehman: Recent Path 3 session-accessor commits and the active tracker show repeated work on the session metadata/accessor migration across plugin SDK, session accessor, runtime, Mattermost, Telegram, and related bundled plugin slices. (role: feature owner and recent area contributor; confidence: high; commits: 4d4769c0d6a5, 4c841ac575d7, b58e6e0734f2; files: src/plugin-sdk/session-store-runtime.ts, src/config/sessions/session-accessor.ts, src/plugins/runtime/runtime-agent.ts)
• Ayaan Zaidi: Blame around the current session-store runtime types and session accessor implementation traces important parts of the accessor seam to earlier grafted work by this author. (role: original seam contributor; confidence: medium; commits: f1e38f2ed6; files: src/plugin-sdk/session-store-runtime.ts, src/config/sessions/session-accessor.ts)
• vincentkoc: Recent history around Codex conversation binding and adjacent channel runtime code makes this owner relevant for the Codex session-metadata read path touched by the PR. (role: adjacent Codex and channel contributor; confidence: medium; commits: 8afc1f770bbe, 5c3469549159, b4cdd9211957; files: extensions/codex/src/conversation-binding.ts, extensions/telegram/src/bot-message-dispatch.ts)
• he-yufeng: Recent Telegram dispatch history overlaps with the internal Telegram metadata reader paths affected by this accessor migration. (role: adjacent Telegram contributor; confidence: low; commits: fa2379dbc883; files: extensions/telegram/src/bot-message-dispatch.ts, extensions/telegram/src/bot-native-commands.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.