Skip to content

fix(session-memory): skip transcript-only assistant rows#95820

Open
moeedahmed wants to merge 2 commits into
openclaw:mainfrom
moeedahmed:fix/session-memory-skip-delivery-mirror
Open

fix(session-memory): skip transcript-only assistant rows#95820
moeedahmed wants to merge 2 commits into
openclaw:mainfrom
moeedahmed:fix/session-memory-skip-delivery-mirror

Conversation

@moeedahmed

@moeedahmed moeedahmed commented Jun 22, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

What Problem This Solves

Session-memory summaries can duplicate assistant replies when a transcript contains both the real provider assistant message and an OpenClaw-authored bookkeeping assistant row with identical visible text, such as delivery-mirror or gateway-injected.

Dropping every OpenClaw-authored assistant row is unsafe, because standalone message-tool delivery mirrors and non-duplicate gateway-injected replies can be the only visible assistant text. This PR now drops only duplicate bookkeeping rows and preserves unique visible assistant replies.

Summary

  • skips OpenClaw delivery-mirror and gateway-injected assistant rows only when their sanitized text repeats the previous visible assistant reply
  • preserves standalone delivery-mirror replies and non-duplicate gateway-injected assistant text
  • resets duplicate comparison on new user turns so later standalone mirrors are not incorrectly dropped
  • adds regression coverage for duplicate dropping and unique-row preservation

Evidence

  • node scripts/run-vitest.mjs run src/hooks/bundled/session-memory/handler.test.ts src/hooks/bundled/session-memory/transcript.test.ts passed: 2 files, 35 tests
  • corepack pnpm exec oxfmt --check --threads=1 src/hooks/bundled/session-memory/transcript.ts src/hooks/bundled/session-memory/handler.test.ts passed
  • git diff --check origin/main...HEAD passed
  • GitHub Real behavior proof check passed on the refreshed head

Real transcript-reader proof using getRecentSessionContent against a temporary JSONL transcript containing:

  • a real provider assistant reply
  • a duplicate delivery-mirror row
  • a duplicate gateway-injected row
  • a non-duplicate gateway-injected row

Output:

user: Summarise the plan
assistant: Plan ready.
assistant: Follow-up sent.

The duplicate OpenClaw-authored bookkeeping rows are omitted, while the unique gateway-injected assistant reply is retained.

Context

Raw transcripts can include both a real assistant response and OpenClaw-authored assistant bookkeeping rows. Session-memory should not duplicate the same assistant text, but it must still preserve OpenClaw-authored assistant rows when they are the only visible reply.

@openclaw-barnacle openclaw-barnacle Bot added size: S triage: needs-pr-context Candidate: external PR body lacks required problem context or evidence. labels Jun 22, 2026
@moeedahmed
moeedahmed force-pushed the fix/session-memory-skip-delivery-mirror branch from 0e3fc4f to e9850db Compare June 22, 2026 16:03
@openclaw-barnacle openclaw-barnacle Bot removed the triage: needs-pr-context Candidate: external PR body lacks required problem context or evidence. label Jun 22, 2026
@clawsweeper

clawsweeper Bot commented Jun 23, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

Codex review: needs maintainer review before merge. Reviewed July 12, 2026, 4:02 AM ET / 08:02 UTC.

Summary
The PR expands session-memory duplicate filtering from delivery-mirror rows to all classified transcript-only OpenClaw assistant rows while preserving unique visible replies and adding regression coverage.

PR surface: Source 0, Tests +112. Total +112 across 2 files.

Reproducibility: yes. Current main only deduplicates delivery-mirror rows, so a provider assistant row followed by an identical gateway-injected row remains a concrete source-level duplicate path.

Review metrics: 1 noteworthy metric.

  • Session-memory predicate: 1 widened. The retention check expands from delivery-mirror-only to both models in the shared transcript-only OpenClaw assistant classification.

Stored data model
Persistent data-model change detected: serialized state: src/hooks/bundled/session-memory/handler.test.ts, serialized state: src/hooks/bundled/session-memory/transcript.ts, unknown-data-model-change: src/hooks/bundled/session-memory/handler.test.ts, unknown-data-model-change: src/hooks/bundled/session-memory/transcript.ts. Confirm migration or upgrade compatibility proof before merge.

Root-cause cluster
Relationship: canonical
Canonical: #95820
Summary: This PR is the open narrow candidate for the gateway-injected remainder of the duplicate session-memory projection bug.

Members:

Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything.

Merge readiness
Overall: 🐚 platinum hermit
Proof: 🦞 diamond lobster
Patch quality: 🐚 platinum hermit
Result: ready for maintainer review.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Rank-up moves:

  • [P2] Refresh or rerun exact-head CI after the unrelated localization baseline repair on current main is included.

Risk before merge

  • [P1] Merging intentionally removes an immediately repeated gateway-injected row from session-memory output, so an unusual workflow that treats identical consecutive injected text as distinct events would lose that event distinction in the memory projection.

Maintainer options:

  1. Accept the narrow projection change (recommended)
    Land the duplicate-only filter after exact-head gates pass, accepting that session memory represents visible reply text rather than repeated bookkeeping events.
  2. Require event-distinction proof
    Pause until a producer-level scenario demonstrates whether identical consecutive gateway-injected rows can represent separate user-relevant events.

Next step before merge

  • No automated repair is needed; this PR should proceed through normal maintainer review and refreshed exact-head merge gates.

Security
Cleared: The diff changes only transcript projection logic and regression tests, with no concrete security or supply-chain concern found.

Review details

Best possible solution:

Keep this PR open until a linked canonical PR proves it covers this PR's unique work, or a maintainer confirms closure.

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

Yes. Current main only deduplicates delivery-mirror rows, so a provider assistant row followed by an identical gateway-injected row remains a concrete source-level duplicate path.

Is this the best way to solve the issue?

Yes. Reusing the shared transcript-only classification while dropping only immediate sanitized duplicates is narrower than general assistant-text deduplication and preserves unique injected replies.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against 55679d88c6be.

Label changes

Label changes:

  • add merge-risk: 🚨 session-state: The PR changes which assistant transcript rows survive into persisted session-memory summaries.
  • add proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes after-fix terminal output from the real transcript reader showing duplicate bookkeeping rows removed and unique gateway-injected text retained.
  • add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
  • add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (terminal): The PR body includes after-fix terminal output from the real transcript reader showing duplicate bookkeeping rows removed and unique gateway-injected text retained.

Label justifications:

  • P2: This is a bounded session-memory correctness fix with limited blast radius.
  • merge-risk: 🚨 session-state: The PR changes which assistant transcript rows survive into persisted session-memory summaries.
  • rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
  • status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (terminal): The PR body includes after-fix terminal output from the real transcript reader showing duplicate bookkeeping rows removed and unique gateway-injected text retained.
  • proof: sufficient: Contributor real behavior proof is sufficient. The PR body includes after-fix terminal output from the real transcript reader showing duplicate bookkeeping rows removed and unique gateway-injected text retained.
Evidence reviewed

PR surface:

Source 0, Tests +112. Total +112 across 2 files.

View PR surface stats
Area Files Added Removed Net
Source 1 8 8 0
Tests 1 112 0 +112
Docs 0 0 0 0
Config 0 0 0 0
Generated 0 0 0 0
Other 0 0 0 0
Total 2 120 8 +112

What I checked:

Likely related people:

  • SunnyShu0925: Authored the merged delivery-mirror duplicate filter that this PR narrowly extends. (role: introduced current behavior; confidence: high; commits: 79a20accca90, 28cbe0024711; files: src/hooks/bundled/session-memory/transcript.ts, src/hooks/bundled/session-memory/handler.test.ts)
  • sallyom: Merged the existing session-memory duplicate fix and is connected to the current projection behavior. (role: merger; confidence: medium; commits: 1052652a7168; files: src/hooks/bundled/session-memory/transcript.ts, src/hooks/bundled/session-memory/handler.test.ts)
  • Peter Steinberger: Introduced the gateway transcript append helper that owns the gateway-injected assistant rows affected by this projection. (role: introduced adjacent behavior; confidence: high; commits: 142c0a7f7d0c; files: src/gateway/server-methods/chat-transcript-inject.ts)
  • vincentkoc: Recently established the shared transcript-only OpenClaw assistant classification reused by this branch. (role: recent area contributor; confidence: medium; commits: b3fb295e9410; files: src/shared/transcript-only-openclaw-assistant.ts)
What the crustacean ranks mean
  • 🦀 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.

How this review workflow works
  • 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.
Review history (2 earlier review cycles)
  • reviewed 2026-07-07T02:43:06.028Z sha 29d2f6c :: needs maintainer review before merge. :: none
  • reviewed 2026-07-07T04:10:08.444Z sha d5ad3e6 :: needs maintainer review before merge. :: none

@clawsweeper clawsweeper Bot added rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. P1 High-priority user-facing bug, regression, or broken workflow. merge-risk: 🚨 session-state 🚨 May lose, corrupt, stale, or mis-associate session, agent, or context state. labels Jun 23, 2026
@moeedahmed
moeedahmed force-pushed the fix/session-memory-skip-delivery-mirror branch from e9850db to d5ad3e6 Compare June 29, 2026 17:27
@moeedahmed

moeedahmed commented Jun 29, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

@clawsweeper

clawsweeper Bot commented Jun 29, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

🦞🧹
ClawSweeper re-review requested.

I asked ClawSweeper to review this item again.
Action: item re-review queued (workflow sweep.yml, event repository_dispatch).
Result: the existing ClawSweeper review comment will be edited in place when the review finishes.

@moeedahmed

moeedahmed commented Jun 29, 2026

Copy link
Copy Markdown
Contributor Author

@clawsweeper re-review

@clawsweeper clawsweeper Bot added proof: sufficient ClawSweeper judged the real behavior proof convincing. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. P2 Normal backlog priority with limited blast radius. and removed rating: 🧂 unranked krab Not merge-ready due to missing proof or serious correctness/safety concerns. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. P1 High-priority user-facing bug, regression, or broken workflow. labels Jun 29, 2026
@clawsweeper clawsweeper Bot mentioned this pull request Jul 1, 2026
Closed
@github-actions github-actions Bot mentioned this pull request Jul 7, 2026
Open
@moeedahmed
moeedahmed force-pushed the fix/session-memory-skip-delivery-mirror branch from 29d2f6c to d5ad3e6 Compare July 7, 2026 03:23
@clawsweeper clawsweeper Bot added proof: sufficient ClawSweeper judged the real behavior proof convincing. and removed proof: sufficient ClawSweeper judged the real behavior proof convincing. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. merge-risk: 🚨 session-state 🚨 May lose, corrupt, stale, or mis-associate session, agent, or context state. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. labels Jul 12, 2026
@clawsweeper clawsweeper Bot added rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. merge-risk: 🚨 session-state 🚨 May lose, corrupt, stale, or mis-associate session, agent, or context state. labels Jul 12, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

merge-risk: 🚨 session-state 🚨 May lose, corrupt, stale, or mis-associate session, agent, or context state. P2 Normal backlog priority with limited blast radius. proof: sufficient ClawSweeper judged the real behavior proof convincing. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. size: S status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR.

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@moeedahmed