Best possible solution:

Keep the conservative exact-prefix, fully anchored ingestion filter and add a redacted terminal capture or runtime log from a controlled Dreaming sweep that shows the corpus output and preserved durable content.

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

Yes at source level: current main accepts normalized session lines after only the minimum-length check, and the linked report supplies concrete assistant-process examples that reach this ingestion path. I did not run a live Dreaming sweep in this read-only review.

Is this the best way to solve the issue?

Yes. A narrowly anchored filter at session-corpus ingestion is the smallest maintainable fix for the reported forward-contamination path, and the branch deliberately preserves durable assistant statements and non-exact lookalikes.

AGENTS.md: unclear because the file could not be read completely.

Codex review notes: model internal, reasoning high; reviewed against ea148e728db3.

Label changes:

• add rating: 🧂 unranked krab: Overall readiness is 🧂 unranked krab; proof is 🧂 unranked krab and patch quality is 🐚 platinum hermit.
• remove rating: 🦪 silver shellfish: Current PR rating is rating: 🧂 unranked krab, so this older rating label is no longer current.
• remove merge-risk: 🚨 session-state: Current PR review selected no merge-risk labels.

Label justifications:

• P2: This is a bounded memory-candidate hygiene fix with limited blast radius and no evidence of an urgent availability, security, or message-delivery failure.
• rating: 🧂 unranked krab: Overall readiness is 🧂 unranked krab; proof is 🧂 unranked krab and patch quality is 🐚 platinum hermit.
• status: 📣 needs proof: The PR needs real behavior proof before ClawSweeper can clear the contributor ask. Needs real behavior proof before merge: The PR body reports focused unit tests, lint, type checks, and CI, but does not include inspectable after-fix output from a real Dreaming session-corpus sweep. Add redacted runtime output or a terminal capture; updating the PR body should trigger a fresh review, or a maintainer can request @clawsweeper re-review. After adding proof, update the PR body; ClawSweeper should re-review automatically. If it does not, the PR author or someone with repository write access can comment @clawsweeper re-review.

PR surface:

Source +13, Tests +143. Total +156 across 2 files.

What I checked:

• Current-main gap: Current main normalizes each session-corpus line and excludes only undersized snippets before recording source provenance and appending candidates; it does not yet reject the reported assistant process-chatter forms. (extensions/memory-core/src/dreaming-phases.ts:635, ea148e728db3)
• Patch placement: The branch adds the predicate immediately after snippet normalization and the minimum-length guard, before line provenance and candidate collection; this is the bounded forward-ingestion boundary identified in the linked report. (extensions/memory-core/src/dreaming-phases.ts:1007, b1638d97cca5)
• Regression scope: The added test exercises rejected chatter alongside preserved durable assistant outcomes and user/lowercase-prefix lookalikes, and asserts corpus source-line and cursor behavior. (extensions/memory-core/src/dreaming-phases.test.ts:1602, b1638d97cca5)
• Problem provenance: The linked report provides concrete observed session-corpus examples and identifies the same normalize-and-ingest hook; it remains open while this candidate fix is unmerged.
• Feature history: Earlier memory-core review history identifies the Dreaming promotion and contamination-guard area as shared work, with vignesh07 introducing the broader Dreaming pipeline and gumadeiras contributing adjacent contamination hardening. (extensions/memory-core/src/dreaming-phases.ts, 4c1022c73b39)

Likely related people:

• vignesh07: Prior area history attributes the Dreaming promotion pipeline and related memory-core surfaces to this contributor. (role: original feature-history owner; confidence: high; commits: 4c1022c73b39; files: extensions/memory-core/src/dreaming-phases.ts, extensions/memory-core/src/short-term-promotion.ts)
• gumadeiras: Prior memory-core history connects this contributor to self-ingestion and contamination hardening that shares the Dreaming corpus invariant. (role: adjacent contamination-guard contributor; confidence: high; commits: 0c4e0d703023; files: extensions/memory-core/src/dreaming-phases.ts, extensions/memory-core/src/short-term-promotion.ts)
• solomonneas: Authored the merged staged-Candidate leak hardening in the neighboring Dreaming promotion path. (role: recent adjacent contributor; confidence: medium; commits: 314903417fcc; files: extensions/memory-core/src/short-term-promotion.ts, extensions/memory-core/src/short-term-promotion.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.

• reviewed 2026-07-20T17:11:55.026Z sha b1638d9 :: needs real behavior proof before merge. :: none