fix(llm): collapse cumulative openai-responses message snapshots instead of concatenating [AI-assisted] by amersheeny · Pull Request #92399 · openclaw/openclaw

amersheeny · 2026-06-12T09:02:33Z

Summary

Fixes #91959 — Bedrock Mantle's openai-responses endpoint (GPT-5.x with reasoning enabled) re-emits the assistant message as many cumulative snapshot items, each a strict prefix-superset of the previous one, instead of a single final message item. The corruption is server-side (confirmed in the issue via raw curl, streaming and non-streaming, usage billed once), but both OpenClaw stream consumers appended one text block per item, so the final visible reply, the session transcript, and the replay context repeated the answer once per snapshot (observed 49–80×).

This adds a generic rule in the shared compat seam (resolveResponsesMessageSnapshotCollapse): a message item whose text strictly extends the immediately preceding text block, with a matching phase, replaces that block (last-wins); identical or shrinking same-phase snapshots are dropped. The first item's textSignature is kept so replay and stream-item identity stay stable. Any non-message output item (reasoning, tool call) resets the tracking, so genuinely distinct post-tool messages (e.g. repeated Codex commentary around tool calls) are never merged and reasoning items are never orphaned for replay. Different-phase (commentary/final_answer) items keep today's separate-block behavior. Applied in the agent transport stream, the shared LLM consumer, and the completed-response backfill; no provider-name keying.

Known scope cut: if a snapshot-emitting provider also re-streams the prefix as output_text.delta events, streaming previews can still transiently show duplicated text mid-stream (downstream phase-aware rendering self-heals at item boundaries); the canonical message, final visible text, transcript, and replay are correct.

AI-assisted (Claude Code), directed and reviewed by the PR author. Session logs available on request.

Real behavior proof

Behavior addressed: A real agent turn against a Responses stream shaped exactly like the corrupted Bedrock Mantle output returned the answer repeated once per snapshot in finalAssistantVisibleText (#91959).

Real environment tested: macOS (Darwin 25.5.0), Node 22.22.1, real CLI end-to-end: openclaw agent --local (embedded agent runner, real openai-responses transport, real session store) on a scratch OPENCLAW_STATE_DIR, with a custom provider (api: "openai-responses", baseUrl: http://127.0.0.1:38291/v1) backed by a local mock server that replays the corrupted wire shape documented in the issue: a reasoning item followed by 4 cumulative message snapshot items (each its own output_item.added / output_text.delta / output_item.done sequence, phase: "final_answer", strictly prefix-increasing), then response.completed with single-copy usage. Live Bedrock Mantle was not reachable from this environment; the wire shape follows the issue's raw SSE captures.

Exact steps or command run after this patch:

node mantle-mock-server.mjs (serves the corrupted SSE stream on 127.0.0.1:38291)
OPENCLAW_STATE_DIR=/tmp/openclaw-91959-proof openclaw agent --local --session-key proof-91959 --model mantlemock/gpt-5.5-mock --json -m "Explain the Transformer self-attention mechanism."
Inspect meta.finalAssistantVisibleText: length and occurrence count of the opening phrase.

Evidence after fix: copied live output — terminal output (stdout) from the real openclaw agent --local runs, inspecting meta.finalAssistantVisibleText of the run's --json result.

Before (same command, same mock stream, the three touched source files from origin/main):

visible text length: 696
opening-phrase occurrences: 4
--- first 400 chars ---
Self-attention projects each token into Q, K and V vectors.
Self-attention projects each token into Q, K and V vectors. Scaled dot-product attention divides QK^T by sqrt(d_k) before the softmax.
Self-attention projects each token into Q, K and V vectors. Scaled dot-product attention divides QK^T by sqrt(d_k) before the softmax. Multi-head attention runs several projections in parallel and concaten

After (this patch):

visible text length: 280
opening-phrase occurrences: 1
--- text ---
Self-attention projects each token into Q, K and V vectors. Scaled dot-product attention divides QK^T by sqrt(d_k) before the softmax. Multi-head attention runs several projections in parallel and concatenates the heads. Positional encodings such as RoPE inject order information.

Observed result after fix: the reply contains the answer a single time; the persisted assistant message holds one text block carrying the first snapshot item's signature.

What was not tested: live Bedrock Mantle (no AWS credentials in this environment — the server-side corruption itself is already evidenced in the issue by raw curl transcripts from @DaiMingNJ); GPT-5.4-shaped variants (replayInvalid=true runs); multi-turn replay against a live provider after a collapsed turn; channel streaming previews during the corrupted stream (covered by code reading only, see scope cut above).

Verification

pnpm test src/llm/** (24 files, 227 tests) and src/agents/openai-transport-stream.test.ts + responses replay/payload/subscriber suites (31 files, ~400 tests): all green after rebase onto origin/main (d481994).
New regression tests: cumulative-snapshot collapse (streaming + backfill, both consumers), repeated-snapshot drop, and negatives pinning the boundaries — different phases, a reasoning item between snapshots, and a tool call between prefix-nested messages must NOT collapse. The collapse tests fail without the fix (verified by reverting the three source files).
pnpm build green; tsgo core + core-test lanes green; oxlint/oxfmt clean; pnpm check:import-cycles green (0 cycles).
pnpm check:changed delegates to Blacksmith Testbox, which is not reachable from this environment (crabbox binary sanity check fails locally) — relying on PR CI for the canonical changed gates.
Local codex review --base origin/main hit the account usage limit (resets 11:41 AM local); a multi-agent adversarial review pass (7 finder angles + verification) was run instead, and its main finding — the first iteration could collapse across reasoning/tool-call boundaries, risking dropped post-tool messages and orphaned reasoning items in replay — is fixed and pinned by the negative tests above.

Release-note context: user-facing fix — replies through openai-responses providers that emit cumulative message snapshots (observed: Bedrock Mantle with GPT-5.x reasoning) no longer repeat the answer dozens of times; transcripts and replay context stay clean. Surface: src/agents/openai-transport-stream.ts, src/llm/providers/openai-responses-shared.ts, src/shared/openai-responses-stream-compat.ts. Reported by @phoenixyy with server-side evidence from @DaiMingNJ in #91959.

clawsweeper · 2026-06-12T09:04:32Z

Codex review: needs maintainer review before merge. Reviewed June 21, 2026, 3:04 PM ET / 19:04 UTC.

Summary
The PR adds a shared strict-prefix cumulative snapshot collapse rule and applies it to the OpenAI Responses shared stream consumer, agent transport stream, completed-response backfill, and regression tests.

PR surface: Source +231, Tests +527. Total +758 across 5 files.

Reproducibility: yes. at source and proof level. The linked issue gives exact CLI and raw-provider evidence, current main still appends each Responses message item independently, and the PR includes after-fix real CLI output against the captured wire shape; I did not rerun live Bedrock Mantle.

Review metrics: 1 noteworthy metric.

Responses Consumption Paths: 2 stream consumers plus 1 completed-response backfill. The same collapse invariant must hold across all three paths or transcripts and replay can diverge by runtime entry point.

Root-cause cluster
Relationship: fixed_by_candidate
Canonical: #91959
Summary: This PR is the active implementation candidate for the canonical cumulative OpenAI Responses snapshot issue; the older overlapping report and earlier unmerged PR are not the current landing path.

Members:

canonical: Bedrock Mantle (openai-responses) cumulatively duplicates reply text with reasoning enabled (GPT-5.x) #91959 - The PR body uses closing linkage for this issue, and the issue contains the stronger raw Bedrock Mantle streaming/non-streaming evidence.
partial_overlap: [Bug] amazon-bedrock-openai / openai-responses: streaming emits multiple incremental final_answer phases, causing duplicate channel messages #89531 - The older closed report described duplicate final-answer output on a related Bedrock openai-responses route, but it is superseded for tracking by the stronger canonical issue and this active PR.
superseded: fix(openai-responses): collapse cumulative message snapshots #92300 - The earlier unmerged PR attempted the same class of mitigation but is closed while this PR remains open and mergeable as the active candidate.

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:

A maintainer acceptance note on the generic strict-prefix heuristic would remove the remaining merge-readiness uncertainty.

Risk before merge

[P1] Merging intentionally treats adjacent same-phase strict-prefix Responses message items as cumulative snapshots across providers; a legitimate provider sequence with that exact shape would lose a visible message boundary in final text, transcript, and replay context.
[P1] The after-fix real behavior proof uses a local server replaying the captured corrupted Bedrock Mantle wire shape, not a live Bedrock Mantle run after this exact head.
[P1] The PR body notes streaming previews can still transiently show duplicated prefixes before item boundaries self-heal the canonical final text.

Maintainer options:

Accept Generic Strict-Prefix Collapse (recommended)
Land with explicit maintainer acceptance that same-phase strict-prefix adjacent Responses message items are snapshots in OpenClaw's supported stream consumers.
Scope The Heuristic
Ask for provider or model gating if maintainers want to preserve generic adjacent message boundaries by default until a stronger marker exists.
Pause For Live Mantle Proof
Wait for a live Bedrock Mantle after-fix run only if maintainers require hosted-provider confirmation beyond the raw captures and local CLI replay.

Next step before merge

[P2] Human review should decide whether the generic strict-prefix snapshot heuristic is an acceptable tradeoff before merge; there is no narrow automated repair to queue.

Security
Cleared: The diff changes runtime stream parsing and tests only; I found no dependency, workflow, script, secret, or supply-chain change that adds a concrete security concern.

Review details

Best possible solution:

Land this PR if maintainers accept the generic strict-prefix message-boundary heuristic; otherwise request a scoped variant or a stronger protocol marker before merging.

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

Yes at source and proof level. The linked issue gives exact CLI and raw-provider evidence, current main still appends each Responses message item independently, and the PR includes after-fix real CLI output against the captured wire shape; I did not rerun live Bedrock Mantle.

Is this the best way to solve the issue?

Yes, with maintainer acceptance of the heuristic. A shared Responses stream/backfill mitigation is the narrow OpenClaw-owned layer, and the PR covers the sibling consumers and boundary negatives instead of adding provider-name-specific duplicate logic.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning high; reviewed against edd1d3319cce.

Label changes

Label justifications:

P1: The PR targets a linked high-impact bug where one assistant turn can duplicate visible output dozens of times and contaminate transcript/replay context.
merge-risk: 🚨 message-delivery: The diff changes how assistant message item boundaries become visible replies, which can suppress duplicate snapshots but could also merge a legitimate same-phase prefix-adjacent message pair.
merge-risk: 🚨 session-state: The diff changes the persisted text block and replay signature for collapsed Responses message items, so transcript and future replay state are directly affected.
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 provides after-fix terminal output from a real openclaw agent --local run against a local server replaying the captured corrupted Responses stream, showing the opening phrase drop from four occurrences to one.
proof: sufficient: Contributor real behavior proof is sufficient. The PR body provides after-fix terminal output from a real openclaw agent --local run against a local server replaying the captured corrupted Responses stream, showing the opening phrase drop from four occurrences to one.

Evidence reviewed

PR surface:

Source +231, Tests +527. Total +758 across 5 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	3	294	63	+231
Tests	2	527	0	+527
Docs	0	0	0	0
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	5	821	63	+758

What I checked:

Current main still appends every shared Responses message item: On current main, the shared Responses stream consumer opens a new text block for each response.output_item.added message and has no snapshot-collapse helper in this path. (src/llm/providers/openai-responses-shared.ts:595, edd1d3319cce)
Current main still appends every completed-response message item: On current main, the agent transport completed-response backfill appends each message item as its own text block, matching the linked issue's cumulative-output failure mode. (src/agents/openai-transport-stream.ts:1482, edd1d3319cce)
PR adds a narrow shared collapse helper: The PR helper collapses only same-phase strict extensions of the immediately preceding message item and keeps equal, shrinking, different-phase, or boundary-separated items distinct. (src/shared/openai-responses-stream-compat.ts:65, 59b3a6efab2b)
PR applies the invariant in both stream consumers: The shared consumer and agent transport defer possible snapshot blocks, replace the prior block on strict extension, preserve the newest item signature, and reset the collapse anchor on non-message boundaries. (src/llm/providers/openai-responses-shared.ts:835, 59b3a6efab2b)
PR covers streaming and backfill regressions: The new tests cover collapse, equal/shrinking negatives, divergence, reasoning/tool boundaries, different phases, balanced event lifecycle, and completed-response backfill. (src/agents/openai-transport-stream.test.ts:438, 59b3a6efab2b)
Downstream text-end delivery already expects growing full snapshots: The subscriber handler treats text_end full-content repeats as suffix-or-noop updates, which supports the PR's same-index repeated text_end shape for grown snapshots. (src/agents/embedded-agent-subscribe.handlers.messages.ts:202, edd1d3319cce)

Likely related people:

vincentkoc: Introduced the LLM transport adapter seam and recently maintained Azure/OpenAI Responses stream behavior in the affected shared and agent transport surfaces. (role: feature-history owner and recent area contributor; confidence: high; commits: 4798e125f40c, 9366d0a873e2, 1727ec7b2dca; files: src/llm/providers/openai-responses-shared.ts, src/agents/openai-transport-stream.ts, src/shared/openai-responses-stream-compat.ts)
steipete: Recently changed OpenAI-compatible reasoning, Responses tool handling, and adjacent agent transport behavior that shares the same message/reasoning boundary concerns. (role: recent adjacent contributor; confidence: medium; commits: 06434d85a053, 5259fa4495ac, 439a9e97fd61; files: src/llm/providers/openai-responses-shared.ts, src/agents/openai-transport-stream.ts)
latensified: Authored recent Responses replay-id work in the same shared and transport paths, relevant to the PR's choice to preserve the newest collapsed message item's signature. (role: recent replay-path contributor; confidence: medium; commits: 6653193fdb90; files: src/llm/providers/openai-responses-shared.ts, src/agents/openai-transport-stream.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.

…em signature Address ClawSweeper P1 review findings on openclaw#92399: text-prefix relation alone was broader than the observed corruption. Equal or shrinking adjacent same-phase message items are now always kept as distinct blocks (the Responses protocol allows multiple message items per response — verified against the sibling Codex parser, codex-rs/codex-api/src/sse/ responses.rs, which emits every output_item.done message as an independent item). With extension-only collapse a false positive can only merge rendering of two messages; it can never remove text. The merged block now carries the newest item's signature instead of the first one's, so replay associates the final content with the item that actually produced it.

amersheeny · 2026-06-12T10:10:27Z

Addressed the review findings in 786c351:

Preserve distinct same-phase message items — the collapse is now extension-only; the drop branch is gone. Equal or shrinking adjacent items are always kept as distinct blocks, with negative tests for both shapes in the streaming path (openai-responses-shared.test.ts) and the completed-response backfill (openai-transport-stream.test.ts). Both raw captures in Bedrock Mantle (openai-responses) cumulatively duplicates reply text with reasoning enabled (GPT-5.x) #91959 are strictly prefix-increasing, so extension-only still fully fixes the corruption — re-verified with the real openclaw agent --local repro against the captured wire shape (280 chars, opening phrase 1×). With extension-only, a false positive can only merge the rendering of two messages into one block; no text can ever be removed.
Replay item identity — the merged block now carries the newest item's signature, so the replayed message item id is the one whose content is byte-identical to the final text.
Personally checked the sibling Codex parser (codex-rs/codex-api/src/sse/responses.rs:280): it does emit every output_item.done message as an independent item. Adjacent same-phase Codex commentary around tool calls is additionally protected by the boundary reset (any reasoning/function_call item clears the tracking).

On the "stronger malformed-sequence marker" option: in the captured corruption the snapshot items carry distinct ids, a shared phase, and re-streamed deltas — none of which discriminates them from legitimate items, so no protocol-level invariant exists beyond the prefix chain itself. Narrowing to strict extension confines the heuristic to the subset where the failure mode is lossless by construction.

@clawsweeper re-review

clawsweeper · 2026-06-12T10:10:31Z

🦞🧹
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.

Re-review progress:

State: Complete
Detail: The targeted re-review finished, the durable review comment was updated, and the synced verdict was routed.
Run: https://github.com/openclaw/clawsweeper/actions/runs/27409262027
Updated: 2026-06-12T10:16:04.699Z

…ifecycle balanced Address the remaining ClawSweeper P1 on openclaw#92399: collapsing a snapshot used to pop a block whose text_start had already been emitted, leaving per-index stream subscribers tracking a phantom block. A message item that follows a finalized text block now defers its public block: no text_start is emitted and deltas are withheld until the item either diverges from the prior text (then the block opens and the withheld prefix replays as one delta) or completes. A collapsed snapshot therefore never starts a block — it only re-ends the prior index with grown content, the documented resend shape — and a distinct deferred item opens and closes its own block normally. No block is ever removed, so every text_start has exactly one matching text_end at a live index. Tests now assert the complete ordered event sequence for the collapse, distinct-item, and divergence cases in both consumers.

amersheeny · 2026-06-12T12:22:59Z

Addressed the remaining P1 in aa8fb1f with the deferred-start shape the review recommended:

Balanced event lifecycle — a message item that follows a finalized text block no longer opens a public block at output_item.added. Its block is deferred: deltas are withheld while the accumulating text is still on the prior block's prefix path, and the block materializes the moment the text diverges (the withheld prefix replays as a single text_delta, then streaming continues live) or at output_item.done for equal/shrinking items. A collapsed snapshot therefore never emits a text_start at all — it only re-ends the prior index with grown content (the existing documented resend shape) — and blocks.pop() is gone from both consumers: no block is ever removed, and every text_start gets exactly one matching text_end at a live index.
Complete event-sequence assertions — the collapse, distinct-item (equal + shrinking), and divergence cases now assert the full ordered [type, contentIndex] event list in both the shared consumer and transport tests (e.g. collapse: start(0), delta(0), end(0), end(0), end(0); distinct: start(0), end(0), start(1), end(1)).
Re-ran both focused test files (20 + 253 passing), the wider llm/agents suites, lint shards, tsgo, build, and the real openclaw agent --local repro against the captured corrupted stream (answer rendered once, 280 chars).

@clawsweeper re-review

clawsweeper · 2026-06-12T12:23:02Z

🦞🧹
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.

Re-review progress:

State: Complete
Detail: The targeted re-review finished, the durable review comment was updated, and the synced verdict was routed.
Run: https://github.com/openclaw/clawsweeper/actions/runs/27493339504
Updated: 2026-06-14T08:47:22.808Z

@phoenixyy

…ead of concatenating Some openai-responses providers (observed: Bedrock Mantle with GPT-5.x reasoning enabled, confirmed server-side via raw curl) re-emit the assistant message as many cumulative snapshot items — each a prefix-superset of the previous one — instead of a single final message item. Both stream consumers appended one text block per item, so the final visible reply, transcript, and replay context repeated the answer once per snapshot (observed 49-80x). Treat a same-phase message item whose text extends the immediately preceding text block as a replacement: the prior block takes the longer text, the duplicate block is dropped, and the first item's signature is kept so replay and stream-item identity stay stable. Shrinking or identical adjacent snapshots are dropped. Any non-message output item (reasoning, tool call) is a real boundary that resets the collapse, so distinct post-tool messages and reasoning replay pairing are untouched, as are different-phase (commentary/final_answer) items. Applies to the agent transport stream, the shared LLM consumer, and completed-response backfill. Fixes openclaw#91959. Reported by @phoenixyy with server-side evidence from @DaiMingNJ.

…em signature Address ClawSweeper P1 review findings on openclaw#92399: text-prefix relation alone was broader than the observed corruption. Equal or shrinking adjacent same-phase message items are now always kept as distinct blocks (the Responses protocol allows multiple message items per response — verified against the sibling Codex parser, codex-rs/codex-api/src/sse/ responses.rs, which emits every output_item.done message as an independent item). With extension-only collapse a false positive can only merge rendering of two messages; it can never remove text. The merged block now carries the newest item's signature instead of the first one's, so replay associates the final content with the item that actually produced it.

…ifecycle balanced Address the remaining ClawSweeper P1 on openclaw#92399: collapsing a snapshot used to pop a block whose text_start had already been emitted, leaving per-index stream subscribers tracking a phantom block. A message item that follows a finalized text block now defers its public block: no text_start is emitted and deltas are withheld until the item either diverges from the prior text (then the block opens and the withheld prefix replays as one delta) or completes. A collapsed snapshot therefore never starts a block — it only re-ends the prior index with grown content, the documented resend shape — and a distinct deferred item opens and closes its own block normally. No block is ever removed, so every text_start has exactly one matching text_end at a live index. Tests now assert the complete ordered event sequence for the collapse, distinct-item, and divergence cases in both consumers.

…ted-response backfill The streaming consumer resets the snapshot-collapse anchor on every non-message output item ("any other item is a real boundary"), but the transport's completed-response backfill only dispatched message and function_call items, so a reasoning item between two strict-prefix message items did not reset the anchor and the later message could collapse across it — an asymmetry with the streaming path's documented invariant. Reset lastTextBlock for every non-message item in the backfill loop (one canonical place; the per-tool-call reset is now redundant and removed). Covered by a backfill reasoning-boundary regression test.

amersheeny · 2026-06-17T08:09:27Z

@clawsweeper re-review

Rebased onto current main (was ~1035 commits behind); new head 59b3a6efab is fully green (143 checks, 0 failures). Same 5 commits, replayed with no conflicts — no behavior change from the SHA last reviewed (4d56b449, which you rated "ready for maintainer review"). Re-requesting a verdict on the current commit.

clawsweeper · 2026-06-17T08:09:31Z

🦞🧹
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.

amersheeny · 2026-06-17T13:32:02Z

@clawsweeper re-review

Retrying — the prior re-review on this SHA failed on Codex capacity, not on the change. Head is green (0 failures) and the diff is identical to the last verdict's commit. Please re-run when a slot is free.

amersheeny · 2026-06-18T07:46:54Z

@clawsweeper re-review

6-hourly retry. Head is green (0 failures) and the diff is unchanged from the last verdict's SHA. Please re-run when an exact-review Codex slot frees up.

amersheeny · 2026-06-18T14:48:33Z

@clawsweeper re-review

The previous run got a Codex slot and the review itself succeeded, but it failed at the "Publish event result and apply safe close" step — so the ack has been stuck at "Review in progress" since ~10:08 UTC with the verdict generated but never posted, and no reschedule followed (publish failures don't re-enter the capacity-retry loop). Re-triggering so the verdict can publish; capacity is free now, so this should complete rather than wait.

@phoenixyy

…ead of concatenating [AI-assisted] (openclaw#92399) * fix(llm): collapse cumulative openai-responses message snapshots instead of concatenating Some openai-responses providers (observed: Bedrock Mantle with GPT-5.x reasoning enabled, confirmed server-side via raw curl) re-emit the assistant message as many cumulative snapshot items — each a prefix-superset of the previous one — instead of a single final message item. Both stream consumers appended one text block per item, so the final visible reply, transcript, and replay context repeated the answer once per snapshot (observed 49-80x). Treat a same-phase message item whose text extends the immediately preceding text block as a replacement: the prior block takes the longer text, the duplicate block is dropped, and the first item's signature is kept so replay and stream-item identity stay stable. Shrinking or identical adjacent snapshots are dropped. Any non-message output item (reasoning, tool call) is a real boundary that resets the collapse, so distinct post-tool messages and reasoning replay pairing are untouched, as are different-phase (commentary/final_answer) items. Applies to the agent transport stream, the shared LLM consumer, and completed-response backfill. Fixes openclaw#91959. Reported by @phoenixyy with server-side evidence from @DaiMingNJ. * test(llm): drop redundant stream drains from responses snapshot tests * fix(llm): collapse only strict snapshot extensions and keep newest item signature Address ClawSweeper P1 review findings on openclaw#92399: text-prefix relation alone was broader than the observed corruption. Equal or shrinking adjacent same-phase message items are now always kept as distinct blocks (the Responses protocol allows multiple message items per response — verified against the sibling Codex parser, codex-rs/codex-api/src/sse/ responses.rs, which emits every output_item.done message as an independent item). With extension-only collapse a false positive can only merge rendering of two messages; it can never remove text. The merged block now carries the newest item's signature instead of the first one's, so replay associates the final content with the item that actually produced it. * fix(llm): defer snapshot-candidate message blocks to keep the event lifecycle balanced Address the remaining ClawSweeper P1 on openclaw#92399: collapsing a snapshot used to pop a block whose text_start had already been emitted, leaving per-index stream subscribers tracking a phantom block. A message item that follows a finalized text block now defers its public block: no text_start is emitted and deltas are withheld until the item either diverges from the prior text (then the block opens and the withheld prefix replays as one delta) or completes. A collapsed snapshot therefore never starts a block — it only re-ends the prior index with grown content, the documented resend shape — and a distinct deferred item opens and closes its own block normally. No block is ever removed, so every text_start has exactly one matching text_end at a live index. Tests now assert the complete ordered event sequence for the collapse, distinct-item, and divergence cases in both consumers. * fix(llm): treat any non-message item as a collapse boundary in completed-response backfill The streaming consumer resets the snapshot-collapse anchor on every non-message output item ("any other item is a real boundary"), but the transport's completed-response backfill only dispatched message and function_call items, so a reasoning item between two strict-prefix message items did not reset the anchor and the later message could collapse across it — an asymmetry with the streaming path's documented invariant. Reset lastTextBlock for every non-message item in the backfill loop (one canonical place; the per-tool-call reset is now redundant and removed). Covered by a backfill reasoning-boundary regression test.

…026.6.11) (#1344) This PR contains the following updates: | Package | Update | Change | |---|---|---| | [ghcr.io/openclaw/openclaw](https://openclaw.ai) ([source](https://github.com/openclaw/openclaw)) | patch | `2026.6.10` → `2026.6.11` | --- ### Release Notes <details> <summary>openclaw/openclaw (ghcr.io/openclaw/openclaw)</summary> ### [`v2026.6.11`](https://github.com/openclaw/openclaw/blob/HEAD/CHANGELOG.md#2026611) [Compare Source](https://github.com/openclaw/openclaw/compare/v2026.6.10...v2026.6.11) We heard the feedback. v2026.6.11 focuses on the rough edges that make OpenClaw feel less dependable, with fixes for misplaced replies, stuck sends, reconnects, model setup failures, and safer admin defaults. [Full release notes](https://docs.openclaw.ai/releases/2026.6.11) ##### Highlights ##### Channel delivery reliability Delivery and reconnect fixes span [Telegram](https://docs.openclaw.ai/channels/telegram), [WhatsApp](https://docs.openclaw.ai/channels/whatsapp), [Matrix](https://docs.openclaw.ai/channels/matrix), [Google Chat](https://docs.openclaw.ai/channels/googlechat), [iMessage](https://docs.openclaw.ai/channels/imessage), [Feishu](https://docs.openclaw.ai/channels/feishu), [Mattermost](https://docs.openclaw.ai/channels/mattermost), [WebChat](https://docs.openclaw.ai/web/webchat), the [Control UI](https://docs.openclaw.ai/web/control-ui), and the [terminal UI](https://docs.openclaw.ai/cli/tui). - Fixes newer Google Chat direct messages sometimes being treated like group conversations, so they reach the correct one-to-one chat while Space and group-chat messages keep their existing routing. [#58993](https://github.com/openclaw/openclaw/pull/58993) Thanks [@starhappysh](https://github.com/starhappysh), [@vincentkoc](https://github.com/vincentkoc). - Feishu voice replies from OpenClaw now show their duration in the chat bubble, so recipients can see how long the audio is before playing it. [#89172](https://github.com/openclaw/openclaw/pull/89172) Related [#53798](https://github.com/openclaw/openclaw/issues/53798). Thanks [@areslp](https://github.com/areslp), [@fxz26284407](https://github.com/fxz26284407), [@kinrocw](https://github.com/kinrocw). - Discord and Telegram replies and mirrored chat history stay tied to the intended conversation more consistently, including across repeated Telegram replies and session changes. [#89911](https://github.com/openclaw/openclaw/pull/89911) Thanks [@jalehman](https://github.com/jalehman). - Background image, video, and music results now return to the chat that requested them when the task starts without a full conversation target, instead of appearing to fail after creation or being sent to the wrong peer as the session moves. [#89949](https://github.com/openclaw/openclaw/pull/89949) Related [#86034](https://github.com/openclaw/openclaw/issues/86034). Thanks [@tianxiaochannel-oss88](https://github.com/tianxiaochannel-oss88), [@wangwllu](https://github.com/wangwllu). - Telegram answers now stay attached to the user's current question when they quote an earlier bot message, while quotes of other people's messages still reply to the selected quote. [#90475](https://github.com/openclaw/openclaw/pull/90475) Thanks [@moeedahmed](https://github.com/moeedahmed). - QQBot group admins can choose how broadly slash commands are available, and private-only commands now direct users to a private chat instead of being exposed or silently ignored in groups. [#92154](https://github.com/openclaw/openclaw/pull/92154) Thanks [@sliverp](https://github.com/sliverp). - Heartbeat checks using reasoning-capable models now show the assistant's intended reply instead of exposing internal reasoning in Telegram, WhatsApp, and other channels, while opt-in Thinking messages still work. [#92356](https://github.com/openclaw/openclaw/pull/92356) Related [#92260](https://github.com/openclaw/openclaw/issues/92260). Thanks [@jmpei](https://github.com/jmpei), [@tangtaizong666](https://github.com/tangtaizong666), [@vincentkoc](https://github.com/vincentkoc). - Telegram progress-mode chats now clear an old progress bubble before newer tool output or artifacts appear, keeping the conversation in a clean, readable order. [#93002](https://github.com/openclaw/openclaw/pull/93002) Related [#90753](https://github.com/openclaw/openclaw/issues/90753). Thanks [@shadow-enthusiast](https://github.com/shadow-enthusiast), [@zhangguiping-xydt](https://github.com/zhangguiping-xydt). - iMessage command-and-link messages now stay together as one OpenClaw turn when delayed link previews arrive, while unrelated quick messages remain separate for users who enabled same-sender DM coalescing. [#93143](https://github.com/openclaw/openclaw/pull/93143) Thanks [@omarshahine](https://github.com/omarshahine). - Successful Discord replies sent through the message tool no longer trigger a misleading failure warning in affected `message_tool_only` source-channel turns. [#94072](https://github.com/openclaw/openclaw/pull/94072) Related [#93875](https://github.com/openclaw/openclaw/issues/93875). Thanks [@chenyangjun-xy](https://github.com/chenyangjun-xy), [@hoyanhan](https://github.com/hoyanhan), [@vincentkoc](https://github.com/vincentkoc). - WhatsApp group conversations now preserve the right message and group context more reliably during retries, reconnects, and group changes. [#94338](https://github.com/openclaw/openclaw/pull/94338) Related [#7433](https://github.com/openclaw/openclaw/issues/7433). Thanks [@mcaxtr](https://github.com/mcaxtr), [@octopuslabs-fl](https://github.com/octopuslabs-fl), [@xialonglee](https://github.com/xialonglee). - Fixes OpenClaw sometimes replying to its own delayed iMessage echoes when stray leading characters keep the sent message from being recognized. [#94442](https://github.com/openclaw/openclaw/pull/94442) Thanks [@ly-wang19](https://github.com/ly-wang19). - Telegram webhook users can keep receiving DMs and group messages through brief channel restarts, configuration reloads, and recovery cycles without temporary message blackouts. [#94506](https://github.com/openclaw/openclaw/pull/94506) Related [#90254](https://github.com/openclaw/openclaw/issues/90254). Thanks [@obviyus](https://github.com/obviyus), [@travellingsoldier85](https://github.com/travellingsoldier85), [@xialonglee](https://github.com/xialonglee). - Matrix E2EE gateways can stay online during long-running use instead of gradually consuming memory until a crash takes channels and in-flight work down. [#94942](https://github.com/openclaw/openclaw/pull/94942) Related [#90455](https://github.com/openclaw/openclaw/issues/90455). Thanks [@xzh-icenter](https://github.com/xzh-icenter), [@yar-sh](https://github.com/yar-sh). - Telegram users now see the intended native reaction instead of leaked instructions or a dropped reaction-only reply, with success recorded only after Telegram accepts it. [#94977](https://github.com/openclaw/openclaw/pull/94977) Related [#71140](https://github.com/openclaw/openclaw/issues/71140). Thanks [@cuttingwater](https://github.com/cuttingwater), [@hugenshen](https://github.com/hugenshen). - Telegram progress updates for commands, searches, updates, and API activity now stay readable instead of exposing noisy HTML or code-style rows, with plain-text fallback when Telegram cannot parse the formatting. [#95007](https://github.com/openclaw/openclaw/pull/95007) Related [#95002](https://github.com/openclaw/openclaw/issues/95002). - Telegram conversations continued in WebChat now show one assistant reply per turn and keep later replies with the active conversation instead of duplicating answers or sending them back to Telegram. [#95069](https://github.com/openclaw/openclaw/pull/95069) Related [#94930](https://github.com/openclaw/openclaw/issues/94930). Thanks [@heichaowo](https://github.com/heichaowo). - Google Chat now hides misleading internal failure banners when a tool result is harmless, leaving users with the completed answer while normal assistant text remains unchanged. [#95084](https://github.com/openclaw/openclaw/pull/95084) Related [#90684](https://github.com/openclaw/openclaw/issues/90684). Thanks [@jailbirt](https://github.com/jailbirt), [@studentzhou-svg](https://github.com/studentzhou-svg). - Bound multi-agent channel conversations now load the workspace files for the configured agent instead of the default agent, though previously misfiled conversations may start fresh in the corrected agent store. [#95118](https://github.com/openclaw/openclaw/pull/95118) Related [#92903](https://github.com/openclaw/openclaw/issues/92903). Thanks [@849261680](https://github.com/849261680), [@axjing](https://github.com/axjing). - People sharing an OpenClaw gateway can now assign different models to individual direct-message contacts across supported chat channels, while existing group and wildcard model choices keep working as before. [#95120](https://github.com/openclaw/openclaw/pull/95120) Related [#53638](https://github.com/openclaw/openclaw/issues/53638). Thanks [@gandalf-at-lerian](https://github.com/gandalf-at-lerian), [@thomaszta](https://github.com/thomaszta), [@xydigit-zt](https://github.com/xydigit-zt). - Telegram now shows that OpenClaw is still working during short initial previews or progress-mode replies instead of leaving the chat silent until the final message arrives. [#95183](https://github.com/openclaw/openclaw/pull/95183) Related [#95004](https://github.com/openclaw/openclaw/issues/95004). Thanks [@obviyus](https://github.com/obviyus). - Matrix users and operators now get a clear failure when a homeserver sends an oversized or stalled response, instead of OpenClaw continuing to buffer it and risking unbounded memory use. [#95240](https://github.com/openclaw/openclaw/pull/95240) Thanks [@alix-007](https://github.com/alix-007). - Fixes delayed or missing Telegram and other queued channel replies in Kubernetes-style deployments with many injected environment variables, where opening the queue database could stall the gateway. [#95278](https://github.com/openclaw/openclaw/pull/95278) Related [#94571](https://github.com/openclaw/openclaw/issues/94571). Thanks [@kaka-srp](https://github.com/kaka-srp). - Telegram chats recover after one stuck message times out, allowing later messages in the same chat or topic to reach the agent without restarting the gateway. [#95299](https://github.com/openclaw/openclaw/pull/95299) Related [#95248](https://github.com/openclaw/openclaw/issues/95248). Thanks [@kriegerbangerz-ship-it](https://github.com/kriegerbangerz-ship-it), [@mikasa0818](https://github.com/mikasa0818), [@obviyus](https://github.com/obviyus). - When people switch between Telegram and another OpenClaw client in a shared direct conversation, short Telegram replies now follow the latest conversation instead of responding to an older, unrelated Telegram proposal. [#95390](https://github.com/openclaw/openclaw/pull/95390) Related [#95378](https://github.com/openclaw/openclaw/issues/95378). Thanks [@maiduy708](https://github.com/maiduy708), [@mikasa0818](https://github.com/mikasa0818), [@obviyus](https://github.com/obviyus). - Fixes completed assistant messages appearing twice in Telegram, Discord, Slack, and other streamed chats after a multi-message reply. [#95432](https://github.com/openclaw/openclaw/pull/95432) Thanks [@vincentkoc](https://github.com/vincentkoc), [@yetval](https://github.com/yetval). - WhatsApp replies now stay attached to the direct or group message being answered instead of appearing as a separate message that loses the conversation context. [#95483](https://github.com/openclaw/openclaw/pull/95483) Thanks [@mcaxtr](https://github.com/mcaxtr). - Telegram rich-message replies now keep paragraphs, bullets, and status lines separated instead of collapsing multi-line content into one run-on block, with no configuration change required. [#95532](https://github.com/openclaw/openclaw/pull/95532) Related [#95409](https://github.com/openclaw/openclaw/issues/95409). Thanks [@amknight](https://github.com/amknight). - Mattermost operators who enable native slash commands can now use `/oc_queue` directly in Mattermost to tune active-run queuing, including its mode, debounce timing, cap, and drop handling. [#95546](https://github.com/openclaw/openclaw/pull/95546) Thanks [@amknight](https://github.com/amknight). - Previously allowed messages keep reaching named accounts after legacy multi-account channel upgrades, with inherited DM and group access rules preserved across Mattermost, Discord, Slack, Telegram, Signal, WhatsApp, iMessage, and IRC. [#95550](https://github.com/openclaw/openclaw/pull/95550) Thanks [@amknight](https://github.com/amknight). - Mattermost users can keep talking in a thread without mentioning the bot again after it replies, and that participation survives gateway restarts until the thread has been idle for seven days. [#95552](https://github.com/openclaw/openclaw/pull/95552) Thanks [@amknight](https://github.com/amknight). - Inbound Telegram messages now reach the configured OpenClaw session promptly instead of sitting unanswered until the next polling interval, a gateway restart, or manual intervention. [#95577](https://github.com/openclaw/openclaw/pull/95577) Related [#86957](https://github.com/openclaw/openclaw/issues/86957). Thanks [@freidrich-goldenflow](https://github.com/freidrich-goldenflow), [@liuwqgit](https://github.com/liuwqgit). - QQBot users now receive complete markdown tables when valid separators use one or two dashes per column, instead of losing the header and all but the final row. [#95637](https://github.com/openclaw/openclaw/pull/95637) Thanks [@ly-wang19](https://github.com/ly-wang19). - Synology Chat users can now receive agent replies that take more than 120 seconds when the configured core timeout allows it, instead of having the channel reject them early. [#95707](https://github.com/openclaw/openclaw/pull/95707) Thanks [@sahibzada-allahyar](https://github.com/sahibzada-allahyar), [@vincentkoc](https://github.com/vincentkoc). - Telegram forum-topic cron jobs now keep separately configured failure alerts going to their intended destination, even when the main announcement uses a topic in the same chat. [#95794](https://github.com/openclaw/openclaw/pull/95794) Thanks [@vincentkoc](https://github.com/vincentkoc). - Fixes WhatsApp group replies that could quote an older OpenClaw message instead of the user's triggering message, so final answers stay attached to the intended message when a reply target is available and avoid pointing back to stale bot context when it is not. [#95914](https://github.com/openclaw/openclaw/pull/95914) Thanks [@mcaxtr](https://github.com/mcaxtr). - WhatsApp users can approve or deny prompts by reaction without the prompt staying stuck when WhatsApp identifies the same direct chat differently, while group approvals remain tied to the correct group and person. [#95935](https://github.com/openclaw/openclaw/pull/95935) Thanks [@mcaxtr](https://github.com/mcaxtr). - Final reply processing now uses less CPU when OpenClaw checks whether block text was already sent, without changing which reply reaches the chat or how duplicate text is suppressed. [#96087](https://github.com/openclaw/openclaw/pull/96087) Thanks [@vincentkoc](https://github.com/vincentkoc). - Exec approval results from external channel plugins now return to the channel or DM where the command started instead of falling back to WebChat or seeming to disappear after approval. [#96140](https://github.com/openclaw/openclaw/pull/96140) Related [#96103](https://github.com/openclaw/openclaw/issues/96103). Thanks [@lansenger-pm](https://github.com/lansenger-pm), [@vincentkoc](https://github.com/vincentkoc), [@yetval](https://github.com/yetval). - WhatsApp's final answer now stays quoted to the follow-up message a user just sent when replying to an older OpenClaw message, instead of arriving unquoted or pointing back to the older bot reply. [#96220](https://github.com/openclaw/openclaw/pull/96220) Thanks [@mcaxtr](https://github.com/mcaxtr). - Nextcloud Talk bots now ignore ordinary file-share and lifecycle events without logging them as bot errors or risking disabled delivery, while malformed chat payloads still return an error. [#96243](https://github.com/openclaw/openclaw/pull/96243) Related [#81566](https://github.com/openclaw/openclaw/issues/81566). Thanks [@arkyu2077](https://github.com/arkyu2077), [@rafaelmgbh](https://github.com/rafaelmgbh), [@vincentkoc](https://github.com/vincentkoc). - Replies and message-tool delivery in Mattermost channels now use channel and thread guidance because the agent identifies those conversations as channels rather than group chats, while existing group-chat behavior remains unchanged. [#96244](https://github.com/openclaw/openclaw/pull/96244) Related [#95645](https://github.com/openclaw/openclaw/issues/95645). Thanks [@arkyu2077](https://github.com/arkyu2077), [@iloveleon19](https://github.com/iloveleon19), [@vincentkoc](https://github.com/vincentkoc). - MCP channel integrations now keep conversation lists, message reads, event polls, and waits within predictable bounds even when a client requests excessive limits or timeouts. [a39e548](https://github.com/openclaw/openclaw/commit/a39e548ede228aa1978bf9d509613cbed6db0c99) Thanks [@vincentkoc](https://github.com/vincentkoc). - Long-running streamed auto-replies are less likely to stop early or abort inconsistently when an unusually large timeout is configured. [6c85b90](https://github.com/openclaw/openclaw/commit/6c85b90469f94955ef00c1609e1f1d6fd2cf4ca8) Thanks [@vincentkoc](https://github.com/vincentkoc). - Channel progress now shows a repeated status when work genuinely returns to it after another update, instead of hiding useful context as a duplicate. [8a75c4d](https://github.com/openclaw/openclaw/commit/8a75c4dd5f3e625a22a7a08c6e1f368798c48111) - Completed channel replies no longer gain late progress notices, preventing stale status text from appearing after the answer is finished. [a594d2c](https://github.com/openclaw/openclaw/commit/a594d2ce73257326b7ab78adb3c4643245ec9431) Thanks [@vincentkoc](https://github.com/vincentkoc). - During streaming channel replies, progress messages now keep showing the latest state instead of getting stuck on an older update. [e114001](https://github.com/openclaw/openclaw/commit/e114001ccafa83b8b366e095a9d7748dfc50c082) Thanks [@vincentkoc](https://github.com/vincentkoc). - Matrix forced resets now handle unavailable secret storage without a runtime error, treating recovery access as unavailable so the reset path can continue safely. [5c5a8a4](https://github.com/openclaw/openclaw/commit/5c5a8a49d76954b53fefc2463bc7b1d6b960e8fc) Thanks [@vincentkoc](https://github.com/vincentkoc). - Configured channels now remain visible in `openclaw channels status --json`, while scheduled announcements reject stale entries that have no active plugin to deliver them. [a641c0d](https://github.com/openclaw/openclaw/commit/a641c0d560fd15373e462829facf15fd6a466aeb) - Discord voice conversations now keep back-to-back assistant responses moving, so a queued reply plays after the previous audio stream closes instead of remaining stuck. [88b64e4](https://github.com/openclaw/openclaw/commit/88b64e4b869e696d99de7417fb52425e9ed67cbf) Thanks [@vincentkoc](https://github.com/vincentkoc). - Discord progress previews are less likely to stop before the final edits when an agent response has already started arriving. [86ea382](https://github.com/openclaw/openclaw/commit/86ea382121b00e73af4b4c329d0a2447592e4071) - Chats no longer show stray `NO_REPLY` text when the assistant means to stay silent, while legitimate media responses still arrive without the placeholder. [96c6f80](https://github.com/openclaw/openclaw/commit/96c6f8022c2420826830b11f4353ce855ab2ac5c) - Telegram streaming replies now show each progress heading once, keeping tool and search updates easier to scan. [013e33c](https://github.com/openclaw/openclaw/commit/013e33c6d3672a980550912442bb1ac5505918aa) Thanks [@vincentkoc](https://github.com/vincentkoc). - Telegram messages that get stuck after a long-running task, crash, or gateway restart now resume processing automatically, so later messages no longer wait silently or require operators to repair the queue by hand. [#97543](https://github.com/openclaw/openclaw/pull/97543) Thanks [@romneyda](https://github.com/romneyda), [@vincentkoc](https://github.com/vincentkoc). ##### Provider and model recovery [Model setup](https://docs.openclaw.ai/providers/models), [OpenAI](https://docs.openclaw.ai/providers/openai), [OpenRouter](https://docs.openclaw.ai/providers/openrouter), [opencode-go](https://docs.openclaw.ai/providers/opencode-go), and [fast-mode](https://docs.openclaw.ai/tools/thinking#fast-mode-fast) follow-up behavior now recover more clearly in affected configurations. - MiniMax text-to-speech and voice notes are less likely to fail because OpenClaw now explicitly requests the audio format it can decode instead of relying on provider defaults. [#73079](https://github.com/openclaw/openclaw/pull/73079) Thanks [@efe-arv](https://github.com/efe-arv). - Gateway operators can again see provider, model, request status, and timing details in normal logs, making model-routing and transport problems easier to diagnose without enabling extra debug logging. [#89648](https://github.com/openclaw/openclaw/pull/89648) Related [#89300](https://github.com/openclaw/openclaw/issues/89300). Thanks [@enominera](https://github.com/enominera), [@xiaobao-k8s](https://github.com/xiaobao-k8s). - Models reached through Google, Mistral, OpenAI Responses, Azure OpenAI Responses, and ChatGPT/Codex Responses now receive clean system instructions without OpenClaw's internal cache-boundary marker leaking into the prompt. [#89716](https://github.com/openclaw/openclaw/pull/89716) Thanks [@enominera](https://github.com/enominera), [@masatohoshino](https://github.com/masatohoshino). - Cron tool calls using Gemini models through OpenAI-compatible providers now run without nullable fields triggering provider schema rejections. [#91559](https://github.com/openclaw/openclaw/pull/91559) Related [#91542](https://github.com/openclaw/openclaw/issues/91542). Thanks [@pick-cat](https://github.com/pick-cat), [@qiukui666](https://github.com/qiukui666). - Provider-qualified model IDs now honor their configured agent runtime policies and CLI aliases instead of unexpectedly falling back to OpenClaw's default runtime. [#91724](https://github.com/openclaw/openclaw/pull/91724) Thanks [@vincentkoc](https://github.com/vincentkoc), [@yu-xin-c](https://github.com/yu-xin-c). - The chat `/models` list and other plugin-aware model or provider selection paths now respond quickly instead of stalling for seconds and consuming a CPU core through repeated setup scans, while plugin changes still refresh normally. [#93356](https://github.com/openclaw/openclaw/pull/93356) Thanks [@obuchowski](https://github.com/obuchowski). - Hosted Ollama Cloud users can keep only the models they explicitly configured after a restart, without the full shared catalog being added back, while automatic discovery continues for local and self-hosted Ollama servers. [#93956](https://github.com/openclaw/openclaw/pull/93956) Thanks [@jason-allen-oneal](https://github.com/jason-allen-oneal). - Cron jobs can now retry or switch to a configured fallback model when a local provider returns the generic `LLM request failed.` error, instead of failing with the fallback unused. [#94062](https://github.com/openclaw/openclaw/pull/94062) Related [#93931](https://github.com/openclaw/openclaw/issues/93931). Thanks [@hugenshen](https://github.com/hugenshen). - Expired provider tokens no longer bury useful operator logs under repeated fallback warnings, while the first warning and later duplicate summaries remain available for diagnosis. [#94233](https://github.com/openclaw/openclaw/pull/94233) Related [#56979](https://github.com/openclaw/openclaw/issues/56979). Thanks [@goutamadwant](https://github.com/goutamadwant), [@yanan1991](https://github.com/yanan1991). - Google Gemini 3.5 Flash can now be selected with its full 1,048,576-token context window, avoiding missing-model errors and needless prompt-size rejections. [#94726](https://github.com/openclaw/openclaw/pull/94726) Related [#94723](https://github.com/openclaw/openclaw/issues/94723). Thanks [@ajwan8998](https://github.com/ajwan8998), [@anguslogan01](https://github.com/anguslogan01), [@kevinat](https://github.com/kevinat). - Dashboard child sessions now handle allowed provider-qualified model choices consistently and give accurate recovery guidance when saved model state is stale. [#94752](https://github.com/openclaw/openclaw/pull/94752) Related [#94713](https://github.com/openclaw/openclaw/issues/94713). Thanks [@gr4via](https://github.com/gr4via). - Claude CLI users no longer get promises of completion updates that may never arrive, because OpenClaw now blocks unsupported native background work before it can strand progress. [#95008](https://github.com/openclaw/openclaw/pull/95008) Thanks [@anagnorisis2peripeteia](https://github.com/anagnorisis2peripeteia). - OpenClaw now rejects oversized provider catalog or JSON responses with a clear error before buffering the entire response in memory. [#95218](https://github.com/openclaw/openclaw/pull/95218) Thanks [@alix-007](https://github.com/alix-007). - OpenRouter users can now select and run the advertised short DeepSeek V4 model IDs without requests failing with `model_not_found` because OpenClaw sent a duplicated provider prefix. [#95268](https://github.com/openclaw/openclaw/pull/95268) Related [#95198](https://github.com/openclaw/openclaw/issues/95198). Thanks [@daniel-alejandro-t](https://github.com/daniel-alejandro-t), [@darren2030](https://github.com/darren2030). - With `/reasoning on`, DeepSeek-style OpenAI-compatible models now show the final answer separately from their reasoning instead of folding it into the reasoning block, with no configuration change required. [#95283](https://github.com/openclaw/openclaw/pull/95283) Related [#95280](https://github.com/openclaw/openclaw/issues/95280). Thanks [@marvinthebored](https://github.com/marvinthebored), [@vincentkoc](https://github.com/vincentkoc), [@zengwen-dt](https://github.com/zengwen-dt). - When a Codex subscription reaches its usage limit, OpenClaw now moves to configured fallback models instead of stopping on the failed result, and it does not retry runs that already produced visible output. [#95400](https://github.com/openclaw/openclaw/pull/95400) Thanks [@jason-allen-oneal](https://github.com/jason-allen-oneal), [@sallyom](https://github.com/sallyom). - LM Studio users can now run quantized or multi-variant local models without false assistant-turn failures or phantom suffixed model entries caused by mismatched model keys. [#95401](https://github.com/openclaw/openclaw/pull/95401) Thanks [@monkeyleet](https://github.com/monkeyleet). - Google-backed embedded-agent runs now stop reading oversized or never-ending prompt-cache responses before they can exhaust memory or leave the run stalled. [#95417](https://github.com/openclaw/openclaw/pull/95417) Thanks [@alix-007](https://github.com/alix-007). - OpenRouter model scans fail safely on oversized or malformed catalogs instead of risking excessive memory use that can destabilize OpenClaw. [#95418](https://github.com/openclaw/openclaw/pull/95418) Thanks [@alix-007](https://github.com/alix-007). - OpenRouter setups now reject oversized model catalogs before they can exhaust OpenClaw's memory, without caching or immediately refetching the failed response. [#95420](https://github.com/openclaw/openclaw/pull/95420) Thanks [@alix-007](https://github.com/alix-007), [@sallyom](https://github.com/sallyom). - Configured fallback models can now answer when Claude CLI runs out of credits or hits a generic runner failure, instead of leaving users with the failure message as the final response. [#95508](https://github.com/openclaw/openclaw/pull/95508) Related [#95489](https://github.com/openclaw/openclaw/issues/95489). Thanks [@mikasa0818](https://github.com/mikasa0818), [@riazrahaman](https://github.com/riazrahaman), [@sallyom](https://github.com/sallyom). - Gemini-backed web searches using `freshness: "day"` or `pd` now complete instead of failing with a provider 400 error, while broader freshness choices and explicit date ranges retain stricter filtering. [#95682](https://github.com/openclaw/openclaw/pull/95682) Thanks [@sunjae-k](https://github.com/sunjae-k), [@vincentkoc](https://github.com/vincentkoc). - Follow-up answers from xAI reasoning models such as Grok Composer now preserve earlier reasoning context more reliably, even when configurable reasoning effort is unsupported. [#95686](https://github.com/openclaw/openclaw/pull/95686) Thanks [@fuller-stack-dev](https://github.com/fuller-stack-dev), [@geraint0923](https://github.com/geraint0923). - Vercel AI Gateway users can now run models chosen from the live catalog, including live-only model IDs that are absent from OpenClaw's bundled list. [#95710](https://github.com/openclaw/openclaw/pull/95710) Thanks [@vincentkoc](https://github.com/vincentkoc). - Fixes manifest-defined providers turning valid model IDs into broken ones when `stripPrefixes` entries have stray spaces or different casing, so operators and plugin authors get the intended provider model. [#95744](https://github.com/openclaw/openclaw/pull/95744) Related [#95743](https://github.com/openclaw/openclaw/issues/95743). Thanks [@parveshsaini](https://github.com/parveshsaini). - First-run setup now opens the credential prompt for a newly installed external provider instead of appearing to loop and leaving OpenAI selected. [#95792](https://github.com/openclaw/openclaw/pull/95792) Related [#95765](https://github.com/openclaw/openclaw/issues/95765). - Oversized or stalled provider catalogs now fail quickly with a clear error instead of hanging OpenClaw or consuming unbounded memory, while normal catalogs continue to load. [#95827](https://github.com/openclaw/openclaw/pull/95827) Thanks [@alix-007](https://github.com/alix-007). - Xiaomi Token Plan users can now use up to 128K output tokens with `mimo-v2.5` and `mimo-v2.5-pro` instead of being stopped at the outdated 32,000-token limit. [#95934](https://github.com/openclaw/openclaw/pull/95934) Thanks [@idootop](https://github.com/idootop). - Tool-heavy model responses can stream with less overhead while repeated tool-call IDs and encrypted reasoning details stay matched to the correct call across Google and OpenAI-compatible providers. [#95957](https://github.com/openclaw/openclaw/pull/95957) Thanks [@vincentkoc](https://github.com/vincentkoc). - Token-usage accounting is more reliable for bundled ACPX users because OpenClaw now includes ACPX 0.11.2's persistence fix by default, without a separate package override or manual client update. [#96124](https://github.com/openclaw/openclaw/pull/96124) Thanks [@vincentkoc](https://github.com/vincentkoc). - Ollama Cloud users can now find and select `glm-5.2:cloud` with its 1,000,000-token context window, reasoning, and tool support even when it is absent from the public model list. [11484f8](https://github.com/openclaw/openclaw/commit/11484f8a1483b7c42aa2971de2d88289fcef7046) - MiniMax image-understanding requests no longer fail before reaching the provider when a timeout is zero, negative, or extremely large; invalid values now use a normal or safe maximum wait. [4b6182e](https://github.com/openclaw/openclaw/commit/4b6182ee2a250005e0c25edfeae4db6ec59b7cb8) Thanks [@vincentkoc](https://github.com/vincentkoc). - Codex runs now follow the current fast-mode choice instead of carrying over an old speed tier, and the status line clearly shows when fast mode is automatic. [77012f9](https://github.com/openclaw/openclaw/commit/77012f9807851c662e064d05097497a25ab13505) Thanks [@vincentkoc](https://github.com/vincentkoc). - Codex-backed conversations now return to normal routing after automatic fast mode is cleared, preventing later turns or model changes from reusing a stale priority tier. [8afc1f7](https://github.com/openclaw/openclaw/commit/8afc1f770bbef30a4d2d9957ef26a685c508448c) - Fallback agent runs now honor each model's configured automatic fast-mode cutoff even when fast mode is overridden for the run, keeping fallback behavior aligned with the selected model policy. [efd3172](https://github.com/openclaw/openclaw/commit/efd3172662ce023eb8d6568b689361536edf06dd) - Live model-switch retries now preserve the original fast-mode cutoff for long-running sessions, while explicit fast mode avoids misleading automatic-cutoff progress messages. [d990115](https://github.com/openclaw/openclaw/commit/d990115d1972fdf4361884a29bbf8396f33e5cba) Thanks [@vincentkoc](https://github.com/vincentkoc). - Embedded agent runs now keep automatic fast mode working consistently through retries and progress updates without confusing it with a manually selected fast-mode setting. [cf1b6fe](https://github.com/openclaw/openclaw/commit/cf1b6fef4403bee7c206299efc4385a7fcb74375) Thanks [@vincentkoc](https://github.com/vincentkoc). - Fast-mode runs now keep their speed setting through model fallback retries and show the configured automatic threshold in status, avoiding inconsistent retry behavior and an unhelpful generic label. [aa3797c](https://github.com/openclaw/openclaw/commit/aa3797c8d0d74b4502d24852ce6baa70286f2f06) Thanks [@vincentkoc](https://github.com/vincentkoc). - Agent replies and scheduled cron runs now handle fast-mode fallback retries more reliably, keeping the state needed for the final attempt to finish or report progress correctly. [14e448e](https://github.com/openclaw/openclaw/commit/14e448e0e13db9f194ea16bb98e0f846a67769fd) Thanks [@vincentkoc](https://github.com/vincentkoc). - Users no longer see a fast-mode reset notice while model fallback attempts are still running; it appears only when the run reaches its final fallback attempt. [6eb72a8](https://github.com/openclaw/openclaw/commit/6eb72a830ece3e2b4c6c85e5a9c2b72b59e0dae9) - Users and operators now get clearer handling when a configured live model becomes unavailable because OpenClaw recognizes the provider's "selected model was not found" response as a model-not-found failure instead of a generic error. [2405d02](https://github.com/openclaw/openclaw/commit/2405d029d437ee58ab94da800a5b213bc6bf4628) Thanks [@vincentkoc](https://github.com/vincentkoc). - Qwen and vLLM now preserve existing chat-template settings consistently when thinking is switched on or off, and provider plugins can use the same tested helper. [2ba9d6e](https://github.com/openclaw/openclaw/commit/2ba9d6eabef9427a950bacc47f077200446cb865) Thanks [@vincentkoc](https://github.com/vincentkoc). - OpenAI-compatible proxy providers can handle thinking levels and legacy `reasoning_effort` fields more consistently, with plugin developers and provider maintainers using one documented normalization helper across OpenRouter, Kilocode, and the SDK. [35bafea](https://github.com/openclaw/openclaw/commit/35bafea757fab0386292951a3dc2a2d3514f370e) - Browser and Vite builds can now load the OpenAI ChatGPT Responses provider without a server-only dependency breaking the bundle, while WebSocket failures still appear normally. [8c8eb86](https://github.com/openclaw/openclaw/commit/8c8eb86fff6e843bd391808ceee249ac8c7f5fa5) Thanks [@vincentkoc](https://github.com/vincentkoc). - OpenRouter model scans now accept the same larger valid catalogs as runtime discovery while still rejecting oversized responses before they can consume unbounded memory. [ad3b2f4](https://github.com/openclaw/openclaw/commit/ad3b2f4b8827cd73b4c1a6c8288569c0966276fe) Thanks [@vincentkoc](https://github.com/vincentkoc). - OpenAI Responses users, including affected Bedrock Mantle GPT-5.x reasoning setups, now get one clean final answer with aligned saved transcripts and replay context instead of dozens of repeated cumulative copies. [#92399](https://github.com/openclaw/openclaw/pull/92399) Related [#91959](https://github.com/openclaw/openclaw/issues/91959). Thanks [@amersheeny](https://github.com/amersheeny), [@daimingnj](https://github.com/daimingnj), [@phoenixyy](https://github.com/phoenixyy), [@pigfoot](https://github.com/pigfoot). - Scheduled jobs and isolated sessions using opencode-go models now move stalled requests into configured timeout or fallback handling instead of hanging for minutes before ending with a generic `LLM request failed` error. [#93965](https://github.com/openclaw/openclaw/pull/93965) Related [#93610](https://github.com/openclaw/openclaw/issues/93610). Thanks [@forceconstant](https://github.com/forceconstant), [@zhangguiping-xydt](https://github.com/zhangguiping-xydt). ##### Session, memory, and trust continuity [Sessions](https://docs.openclaw.ai/concepts/session), [compaction](https://docs.openclaw.ai/concepts/compaction), [memory](https://docs.openclaw.ai/concepts/memory), [QMD-backed memory](https://docs.openclaw.ai/concepts/memory-qmd), and [Tool Search](https://docs.openclaw.ai/tools/tool-search) retain useful state more consistently, while [Matrix](https://docs.openclaw.ai/channels/matrix) recovery, [tool policies](https://docs.openclaw.ai/gateway/sandbox-vs-tool-policy-vs-elevated), and [approvals](https://docs.openclaw.ai/cli/approvals) stay attached to the intended trust boundary. - Affected agent conversations using OpenAI Responses can now recover and keep replying after a visible channel response leaves their saved history incomplete, instead of every later turn failing before a reply appears. [#84708](https://github.com/openclaw/openclaw/pull/84708) Thanks [@anyech](https://github.com/anyech). - When a Codex-backed agent produces unusually large tool output, saved and replayed conversations now keep its text within the usual size limit while leaving non-text content unchanged. [#87912](https://github.com/openclaw/openclaw/pull/87912) Thanks [@adrianip0204](https://github.com/adrianip0204). - Control UI conversations now stay visible and continue in the same session after a sleep, network drop, or Gateway reconnect instead of disappearing when the next message is sent. [#89017](https://github.com/openclaw/openclaw/pull/89017) Related [#87700](https://github.com/openclaw/openclaw/issues/87700). Thanks [@zhangguiping-xydt](https://github.com/zhangguiping-xydt), [@asicoe](https://github.com/asicoe). - Bundled Codex and Copilot integrations now keep mirrored chat history and transcript updates tied to the correct OpenClaw session as storage evolves, while existing file-backed active transcripts continue working during the migration. [#89518](https://github.com/openclaw/openclaw/pull/89518) Thanks [@jalehman](https://github.com/jalehman). - WebChat's current-session status now matches the conversation you are actually using, so the session identity, thinking level, token context, and cost details no longer come from the fallback `main` session. [#89800](https://github.com/openclaw/openclaw/pull/89800) Related [#89773](https://github.com/openclaw/openclaw/issues/89773). Thanks [@killo3967](https://github.com/killo3967), [@sweetcornna](https://github.com/sweetcornna). - Your conversation is less likely to lose its context after you press stop during automatic compaction because the compaction request is now cancelled too. [#89886](https://github.com/openclaw/openclaw/pull/89886) Related [#89868](https://github.com/openclaw/openclaw/issues/89868). Thanks [@lykeion-dev](https://github.com/lykeion-dev), [@openperf](https://github.com/openperf), [@vincentkoc](https://github.com/vincentkoc). - When cross-agent session access is blocked, OpenClaw now lists all required visibility, agent-to-agent, and allow-list settings, helping operators correct policy configuration instead of chasing a nonexistent agent failure. [#90489](https://github.com/openclaw/openclaw/pull/90489) Related [#90443](https://github.com/openclaw/openclaw/issues/90443). Thanks [@ramitrkar-hash](https://github.com/ramitrkar-hash), [@sahibzada-allahyar](https://github.com/sahibzada-allahyar), [@vincentkoc](https://github.com/vincentkoc). - `openclaw memory status` now shows an active light or REM dreaming phase instead of incorrectly reporting `Dreaming: off`, so operators can see that valid memory configurations are enabled. [#93113](https://github.com/openclaw/openclaw/pull/93113) Related [#67868](https://github.com/openclaw/openclaw/issues/67868). Thanks [@agentarclab](https://github.com/agentarclab), [@mrossit](https://github.com/mrossit). - Timed-out QMD memory searches now stop their background work when the agent moves on, preventing abandoned processes from continuing to consume CPU and memory. [#93394](https://github.com/openclaw/openclaw/pull/93394) Thanks [@alix-007](https://github.com/alix-007). - Repeated instructions sent after compaction now remain in the conversation, preventing lost turns, orphaned replies, and malformed history that some providers reject. [#94328](https://github.com/openclaw/openclaw/pull/94328) Thanks [@vincentkoc](https://github.com/vincentkoc), [@yetval](https://github.com/yetval). - Memory Wiki's Stale Pages report now leaves durable concept and synthesis pages out of freshness warnings, keeping attention on source and entity pages that may actually need review. [#94369](https://github.com/openclaw/openclaw/pull/94369) Thanks [@sunnyshu0925](https://github.com/sunnyshu0925), [@vincentkoc](https://github.com/vincentkoc). - Long embedded runs with recent progress are now less likely to be interrupted by stale-session recovery, while genuinely stalled runs can still be cleared so queued work continues. [#94701](https://github.com/openclaw/openclaw/pull/94701) Thanks [@imadal1n](https://github.com/imadal1n), [@mrclawfield](https://github.com/mrclawfield). - Ollama memory search now respects a configured smaller embedding dimension and keeps indexes for different dimensions separate, avoiding incompatible vectors being mixed together. [#94811](https://github.com/openclaw/openclaw/pull/94811) Thanks [@mushuiyu886](https://github.com/mushuiyu886). - Memory searches and targeted refreshes now stay connected to the correct OpenClaw session even when transcript filenames change or QMD exports use a different name. [#95087](https://github.com/openclaw/openclaw/pull/95087) Thanks [@jalehman](https://github.com/jalehman). - Long-running conversations with screenshots or other images now keep their continuity more consistently when OpenClaw makes room for new messages, instead of repeatedly filling up without moving the retained conversation forward. [#95128](https://github.com/openclaw/openclaw/pull/95128) Thanks [@yetval](https://github.com/yetval). - Windows users can now run QMD-backed memory indexing and search through configured absolute `memory.qmd.command` paths, including drive-letter and UNC locations, without OpenClaw stripping the path separators before launch. [#95274](https://github.com/openclaw/openclaw/pull/95274) Related [#92302](https://github.com/openclaw/openclaw/issues/92302). Thanks [@ardooken](https://github.com/ardooken), [@ly85206559](https://github.com/ly85206559). - Usage footers selected with `/usage full` or `/usage tokens` now remain visible after daily or idle session rollover, so users do not have to turn them on again. [#95322](https://github.com/openclaw/openclaw/pull/95322) Thanks [@litang9](https://github.com/litang9). - Follow-up replies, reactions, threaded messages, and status checks stay with the chat they belong to after webchat or system activity, while real channel switches still clear outdated routing details. [#95467](https://github.com/openclaw/openclaw/pull/95467) Thanks [@yetval](https://github.com/yetval). - Long-running main conversations now keep their prior context when users return after an overnight or delayed follow-up, rather than silently starting over after an otherwise normal completion. [#95472](https://github.com/openclaw/openclaw/pull/95472) Thanks [@xydt-tanshanshan](https://github.com/xydt-tanshanshan). - People with large session histories can list, preview, and find sessions without multi-second freezes, while older mixed-case session keys are still migrated at startup. [#95699](https://github.com/openclaw/openclaw/pull/95699) Thanks [@jalehman](https://github.com/jalehman), [@jzakirov](https://github.com/jzakirov). - Fixes delivered replies sometimes being saved to the wrong conversation history, or omitted from it, when operators use a custom or per-agent `session.store`, improving continuity and auditability for the intended session. [#95782](https://github.com/openclaw/openclaw/pull/95782) Related [#95781](https://github.com/openclaw/openclaw/issues/95781). Thanks [@youngting520](https://github.com/youngting520). - Saved session-memory summaries now leave out raw model tokens, tool-call blocks, media placeholders, role tags, and stale `NO_REPLY` markers so future conversations keep useful context. [#95791](https://github.com/openclaw/openclaw/pull/95791) Thanks [@sweetsophia](https://github.com/sweetsophia), [@vincentkoc](https://github.com/vincentkoc), [@yb0y](https://github.com/yb0y). - Long-running OpenAI sessions using Codex/ChatGPT OAuth can now compact without a separate API key, whether `/compact` is run manually or triggered automatically. [#95831](https://github.com/openclaw/openclaw/pull/95831) Related [#95693](https://github.com/openclaw/openclaw/issues/95693). Thanks [@sallyom](https://github.com/sallyom), [@yui-tien](https://github.com/yui-tien). - Long, tool-heavy sessions now compact oversized conversations instead of getting stuck when a large tool result appears at the end. [#95860](https://github.com/openclaw/openclaw/pull/95860) Related [#78478](https://github.com/openclaw/openclaw/issues/78478). Thanks [@jw8957](https://github.com/jw8957), [@wzhgba](https://github.com/wzhgba), [@yetval](https://github.com/yetval). - When `memory_search` is unavailable because the Node runtime lacks `node:sqlite`, OpenClaw now points users to a compatible runtime instead of sending them through unrelated embedding-provider troubleshooting. [#95916](https://github.com/openclaw/openclaw/pull/95916) Thanks [@rrrrrredy](https://github.com/rrrrrredy), [@vincentkoc](https://github.com/vincentkoc). - Developers and operators inspecting a compacted Copilot session now get its summary, before-and-after token counts, and session details instead of an incomplete result. [#96049](https://github.com/openclaw/openclaw/pull/96049) Thanks [@vincentkoc](https://github.com/vincentkoc). - The `/stop` and abort commands now keep stopping active runs, clearing queued followups, and ending related subagents promptly even when session keys need canonicalizing or abort metadata cannot be saved. [#96201](https://github.com/openclaw/openclaw/pull/96201) Thanks [@jalehman](https://github.com/jalehman). - Voice Wake upgrades now keep existing trigger phrases and routing rules working as OpenClaw moves them from retired settings files into the shared state database. [bdf81a8](https://github.com/openclaw/openclaw/commit/bdf81a825fa3ef66ad2c535c1eeb0bb4e31b6d1b) - Upgrades from older OpenClaw state layouts now preserve update notifications, check throttling, available-version records, and automatic-update attempt history as that state moves into SQLite. [eb00d49](https://github.com/openclaw/openclaw/commit/eb00d499d16feea600fceef92d575fa30f005649) Thanks [@vincentkoc](https://github.com/vincentkoc). - Plugin-channel conversations keep their intended session more reliably through startup, doctor checks, and state repairs, with older binding records migrated into OpenClaw's shared database. [9f888d9](https://github.com/openclaw/openclaw/commit/9f888d95e082d50380a66db18ee2e32683e688e0) - Windows memory-backed session syncing now keeps using the intended transcript file even when path formatting differs. [b3b5b08](https://github.com/openclaw/openclaw/commit/b3b5b08e67a26efd648c7c7d879e5487223cd796) Thanks [@vincentkoc](https://github.com/vincentkoc). - Embedded agent runs with a missing or blank session key now stay attached to the intended session instead of being sent through inconsistent session routing. [911f853](https://github.com/openclaw/openclaw/commit/911f853b7fc4d819e2175b001662a01eba30453d) Thanks [@vincentkoc](https://github.com/vincentkoc). - When a model guesses the wrong tool name, Tool Search and Code Mode now show how to find and retry the correct tool, reducing the risk that long-running sessions get stuck or lose durable memory during compaction. [#93374](https://github.com/openclaw/openclaw/pull/93374) Related [#92273](https://github.com/openclaw/openclaw/issues/92273). Thanks [@mushuiyu886](https://github.com/mushuiyu886), [@poison](https://github.com/poison), [@vincentkoc](https://github.com/vincentkoc). - Fixes assistant replies disappearing from webchat, Control UI, Feishu, and other embedded conversations after compaction, keeping refreshed chats readable and follow-up requests separate. [#95484](https://github.com/openclaw/openclaw/pull/95484) Related [#76729](https://github.com/openclaw/openclaw/issues/76729). Thanks [@maweibin](https://github.com/maweibin), [@njuboy11](https://github.com/njuboy11), [@vincentkoc](https://github.com/vincentkoc). - OpenClaw memory features now keep active, reset, and deleted transcript coverage aligned with configured session stores and agent ownership, making dreaming, QMD exports, indexing, and sync less likely to miss or misattribute conversation history. [#96162](https://github.com/openclaw/openclaw/pull/96162) Thanks [@jalehman](https://github.com/jalehman). - Gateway TLS setup now rejects blank certificate or key paths clearly or uses OpenClaw's defaults, avoiding confusing startup and certificate-generation failures while preserving valid paths. [#94054](https://github.com/openclaw/openclaw/pull/94054) Thanks [@miorbnli](https://github.com/miorbnli). - Configured plugin policies keep blocking or rewriting sensitive tool calls after Gateway registry changes, reloads, or later hook initialization instead of being silently skipped. [#94545](https://github.com/openclaw/openclaw/pull/94545) Thanks [@jesse-merhi](https://github.com/jesse-merhi). - Mobile operators with `operator.approvals` can now see and resolve chat-triggered exec approvals on the iOS device that started the request, including while the app is open, without relying only on push notifications. [#95175](https://github.com/openclaw/openclaw/pull/95175) Thanks [@joshavant](https://github.com/joshavant). - Control UI users now get the patched DOMPurify release, reducing exposure to the GHSA-cmwh-pvxp-8882 sanitizer vulnerability without changing how the interface behaves. [#95691](https://github.com/openclaw/openclaw/pull/95691) Thanks [@vincentkoc](https://github.com/vincentkoc). - "Always allow" approvals for plugin conversation bindings now carry over from the old settings file and are less likely to be lost or overwritten when multiple OpenClaw processes are running. [ae41b00](https://github.com/openclaw/openclaw/commit/ae41b009224b0a8e3a990912503258d4478fb4d0) Thanks [@vincentkoc](https://github.com/vincentkoc). - Matrix users now see that the active recovery key is required before a forced cross-signing reset can proceed, preventing a second reset from leaving encryption recovery and room-key backups unusable. [#95720](https://github.com/openclaw/openclaw/pull/95720) Related [#78396](https://github.com/openclaw/openclaw/issues/78396). Thanks [@jteddy](https://github.com/jteddy), [@vincentkoc](https://github.com/vincentkoc), [@xialonglee](https://github.com/xialonglee). ##### Slack router relay mode [Slack router relay mode](https://docs.openclaw.ai/channels/slack) gives managed and multi-gateway deployments a supported way to centralize incoming Slack traffic while preserving gateway ownership of mentions, threads, and replies. - Managed Slack deployments can now use a central router to send mentions and ongoing threads to the right OpenClaw gateway while replies still appear through Slack. [#94707](https://github.com/openclaw/openclaw/pull/94707) Thanks [@pash-openai](https://github.com/pash-openai), [@sjf-oa](https://github.com/sjf-oa). ##### Raft External Agent wake bridge The [Raft channel](https://docs.openclaw.ai/channels/raft) and [Raft plugin](https://docs.openclaw.ai/plugins/reference/raft) now support the local CLI wake path for External Agents. - Raft External Agent operators can now wake an OpenClaw agent when a workspace has pending work through the supported local CLI bridge, with named profiles and checks for missing CLI prerequisites. [#95497](https://github.com/openclaw/openclaw/pull/95497) Thanks [@vincentkoc](https://github.com/vincentkoc). ##### Official plugin installation and repair [Plugin management](https://docs.openclaw.ai/plugins/manage-plugins) and the [plugin inventory](https://docs.openclaw.ai/plugins/plugin-inventory) now cover more official integrations through normal installation, update, and repair workflows. - When `plugins.allow` uses a channel or package name instead of the real plugin id, startup guidance now identifies the unmatched entry and shows the discovered plugin ids needed to correct the configuration. [#68389](https://github.com/openclaw/openclaw/pull/68389) Related [#68352](https://github.com/openclaw/openclaw/issues/68352). Thanks [@aym9999](https://github.com/aym9999), [@jirboy](https://github.com/jirboy), [@lyfuci](https://github.com/lyfuci), [@pahuchi-joe](https://github.com/pahuchi-joe), [@zmxccxy](https://github.com/zmxccxy). - Plugin trust warnings for first-time or fresh installs now include a ready-to-copy `plugins.allow` example and commands to list or inspect plugin ids, so users can resolve the warning before trusting or reinstalling plugin code. [#78105](https://github.com/openclaw/openclaw/pull/78105) Related [#68780](https://github.com/openclaw/openclaw/issues/68780). Thanks [@jirboy](https://github.com/jirboy), [@pahuchi-joe](https://github.com/pahuchi-joe). - Codex migrations now work with standard global plugin installs because `openclaw migrate` can find the installed provider instead of failing with `Unknown migration provider`. [#89612](https://github.com/openclaw/openclaw/pull/89612) Related [#89609](https://github.com/openclaw/openclaw/issues/89609). Thanks [@mugabuga](https://github.com/mugabuga), [@zerone0x](https://github.com/zerone0x). - Plugin installs and updates recover from stale OpenClaw-managed dependency pins instead of failing with `npm EOVERRIDE`, without later synchronization downgrading or removing packages users installed explicitly. [#91786](https://github.com/openclaw/openclaw/pull/91786) Related [#91772](https://github.com/openclaw/openclaw/issues/91772). Thanks [@amknight](https://github.com/amknight), [@mkdelta221](https://github.com/mkdelta221). - Channel plugin developers can now carry native sender and conversation identifiers through hooks and selected exec workflows, giving integrations more precise routing without breaking existing sender and chat fields. [#91903](https://github.com/openclaw/openclaw/pull/91903) Thanks [@lanzhi-lee](https://github.com/lanzhi-lee), [@vincentkoc](https://github.com/vincentkoc). - Plugin discovery now repeats fewer blocking filesystem checks during startup, reducing avoidable cold-start work for bundled plugin trees, especially on slower Windows filesystems, without changing bundle discovery behavior. [#93919](https://github.com/openclaw/openclaw/pull/93919) Related [#76209](https://github.com/openclaw/openclaw/issues/76209). Thanks [@ml12580](https://github.com/ml12580), [@shenhonglong456-ai](https://github.com/shenhonglong456-ai). - Plugin Gateway methods now work through `openclaw gateway call` after registration, so plugin authors can use them from scripts and cron jobs instead of hitting an `unknown method` error. [#94154](https://github.com/openclaw/openclaw/pull/94154) Related [#94127](https://github.com/openclaw/openclaw/issues/94127). Thanks [@brycemurray](https://github.com/brycemurray), [@pick-cat](https://github.com/pick-cat), [@vincentkoc](https://github.com/vincentkoc). - ClawHub skill discovery and install checks are less likely to stall or crash OpenClaw because oversized or stalled marketplace responses are now stopped before they can exhaust memory. [#95226](https://github.com/openclaw/openclaw/pull/95226) Thanks [@alix-007](https://github.com/alix-007). - Pinned official plugins no longer stay on an old release when operators follow the repair advice from `openclaw doctor` or deep gateway status after an upgrade. [#95541](https://github.com/openclaw/openclaw/pull/95541) Thanks [@ooiuuii](https://github.com/ooiuuii), [@vincentkoc](https://github.com/vincentkoc). - Managed npm plugin updates are less likely to break work on a running gateway with missing-module errors, because the older plugin files remain available until a later gateway start cleans them up. [#95589](https://github.com/openclaw/openclaw/pull/95589) Thanks [@ooiuuii](https://github.com/ooiuuii), [@vincentkoc](https://github.com/vincentkoc). - Official plugin cards for supported brands now show recognizable icons in ClawHub and other catalogs, and plugin authors can provide marketplace artwork through the documented manifest field. [#95845](https://github.com/openclaw/openclaw/pull/95845) Thanks [@patrick-erichsen](https://github.com/patrick-erichsen). - Official plugin icons in ClawHub and other catalogs are no longer forced into the same hard-coded color, allowing Simple Icons to use its default artwork instead. [#95987](https://github.com/openclaw/openclaw/pull/95987) Thanks [@patrick-erichsen](https://github.com/patrick-erichsen). - Docker users now have an official `openclaw/openclaw` Docker Hub mirror alongside GHCR, with versioned beta releases kept from moving the stable `latest` and `main` aliases. [#97122](https://github.com/openclaw/openclaw/pull/97122) Thanks [@vincentkoc](https://github.com/vincentkoc). - Git-based OpenClaw installs now use the repository's pinned pnpm version even when another global pnpm or surrounding project package manager is present, so setup commands no longer run against the wrong package-manager environment. [bd74a62](https://github.com/openclaw/openclaw/commit/bd74a62118aa4774706359d9494116ded8c1f6e3) Thanks [@vincentkoc](https://github.com/vincentkoc). - ClawHub skill-card and update requests now complete or time out predictably even when they receive an unusually large timeout value. [8cd0c11](https://github.com/openclaw/openclaw/commit/8cd0c11227f6f4096d089cd6108d6f2ae31252b7) Thanks [@vincentkoc](https://github.com/vincentkoc). - Windows users can complete source installs without a llama.cpp setup step blocking or slowing them, and the installer restores their existing shell setting afterward. [ea9065b](https://github.com/openclaw/openclaw/commit/ea9065bc68dd4ff94495b85a7dcb4491cf41b67a) Thanks [@vincentkoc](https://github.com/vincentkoc). - More official channel, provider, and web-search plugins can now be installed or repaired through normal external package catalogs while still being recognized from their existing credentials. [#95683](https://github.com/openclaw/openclaw/pull/95683) Thanks [@vincentkoc](https://github.com/vincentkoc). ##### Channels and Messaging ##### Additional channel fixes Additional [Telegram](https://docs.openclaw.ai/channels/telegram) and channel configuration fixes cover narrower delivery and setup problems. - Telegram reply chains keep cached replies attached after context changes instead of failing when those cached replies are reused. [#82909](https://github.com/openclaw/openclaw/pull/82909) Thanks [@lidge-jun](https://github.com/lidge-jun). - Fixes Discord dropping an entire long reply with fenced code blocks when a closing code fence lands near the 2,000-character message limit. [#95661](https://github.com/openclaw/openclaw/pull/95661) Thanks [@ly-wang19](https://github.com/ly-wang19). - Slack operators can now store tokens and signing secrets as supported SecretRef inputs, while reads, writes, allowlist and target lookups, and setup checks use the resolved credentials instead of rejecting or misreading the references. [7da955f](https://github.com/openclaw/openclaw/commit/7da955fae4ca2083599aa33a1f93dbfff53cb187) Thanks [@vincentkoc](https://github.com/vincentkoc). - Channel capability checks now return a clear timeout when an integration stops responding, keeping troubleshooting from hanging in a terminal or automation run. [8ecdb97](https://github.com/openclaw/openclaw/commit/8ecdb97b636e4c3fcc6d142d217327404ae06581) Thanks [@vincentkoc](https://github.com/vincentkoc). ##### Providers and Models ##### Additional provider and model fixes Additional [model configuration](https://docs.openclaw.ai/providers/models) fixes improve selection, usage reporting, status output, and streaming. - After changing the default model, starting a fresh channel session with `/new` or `/reset` now uses the new default instead of silently reusing the previous cached model, while explicit `/model` overrides remain unchanged. [#77339](https://github.com/openclaw/openclaw/pull/77339) Related [#77322](https://github.com/openclaw/openclaw/issues/77322). Thanks [@mjamiv](https://github.com/mjamiv), [@zaynl](https://github.com/zaynl). - Behind HTTP or HTTPS proxies, Codex/OpenAI usage and quota checks in `openclaw status --usage --json` and the Control UI now retrieve usage windows instead of failing when chatgpt.com is unreachable directly. [#93943](https://github.com/openclaw/openclaw/pull/93943) Related [#78714](https://github.com/openclaw/openclaw/issues/78714). Thanks [@tnzgit](https://github.com/tnzgit), [@turbotheturtle](https://github.com/turbotheturtle). - `/status` now keeps the active model and how to clear a pinned choice on one compact line, so Discord and other chat users can scan model status without a multi-line explanation. [#95797](https://github.com/openclaw/openclaw/pull/95797) Thanks [@solvely-colin](https://github.com/solvely-colin). - Anthropic streaming responses now keep interleaved text, thinking, and tool-call updates …

openclaw-barnacle Bot added agents Agent runtime and tooling size: L triage: mock-only-proof Candidate: PR proof only shows tests, mocks, snapshots, lint, typecheck, or CI. labels Jun 12, 2026

clawsweeper Bot added rating: 🦪 silver shellfish Thin PR readiness signal; proof, validation, or implementation needs work. status: 📣 needs proof The PR needs real behavior proof before ClawSweeper can clear the contributor ask. labels Jun 12, 2026

openclaw-barnacle Bot added proof: supplied External PR includes structured after-fix real behavior proof. and removed triage: mock-only-proof Candidate: PR proof only shows tests, mocks, snapshots, lint, typecheck, or CI. labels Jun 12, 2026

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 12, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 12, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 14, 2026

amersheeny added 5 commits June 17, 2026 08:27

test(llm): drop redundant stream drains from responses snapshot tests

67413d2

amersheeny force-pushed the fix/openai-responses-cumulative-snapshots branch from 4d56b44 to 59b3a6e Compare June 17, 2026 07:30

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label Jun 17, 2026

clawsweeper Bot removed 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. labels Jun 17, 2026

amersheeny mentioned this pull request Jun 18, 2026

fix(cron): bound restart catch-up to schedule activation #92398

Closed

openclaw-clownfish Bot merged commit 9fbc8a7 into openclaw:main Jun 22, 2026
196 of 202 checks passed

clawsweeper Bot mentioned this pull request Jun 30, 2026

fix(openai-transport): 120-second timeout in OpenAI Responses API streaming #98244

Closed

github-actions Bot mentioned this pull request Jul 1, 2026

chore: bump openclaw/openclaw version from v2026.6.10 to v2026.6.11 Docker-Hub-sirmark/docker-openclaw#35

Merged

github-actions Bot mentioned this pull request Jul 6, 2026

chore(openclaw): bump pinned version v2026.5.28 → v2026.6.11 Evans-Software-Solutions-Limited/axel-saas#121

Open

clawsweeper Bot mentioned this pull request Jul 6, 2026

[Bug]: OpenAI streaming provider fails to map refusal chunks to ModelRefusalError #101059

Open

SunnyShu0925 mentioned this pull request Jul 14, 2026

fix(providers): add itemId check to message snapshot collapse #107831

Closed

23 tasks

clawsweeper Bot mentioned this pull request Jul 17, 2026

fix: preserve OpenAI Responses message boundaries #110059

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(llm): collapse cumulative openai-responses message snapshots instead of concatenating [AI-assisted]#92399

fix(llm): collapse cumulative openai-responses message snapshots instead of concatenating [AI-assisted]#92399
openclaw-clownfish[bot] merged 5 commits into
openclaw:mainfrom
amersheeny:fix/openai-responses-cumulative-snapshots

amersheeny commented Jun 12, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

Uh oh!

amersheeny commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

Uh oh!

amersheeny commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

Uh oh!

amersheeny commented Jun 17, 2026

Uh oh!

clawsweeper Bot commented Jun 17, 2026 •

edited

Loading

Uh oh!

amersheeny commented Jun 17, 2026

Uh oh!

amersheeny commented Jun 18, 2026

Uh oh!

amersheeny commented Jun 18, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Uh oh!

Conversation

amersheeny commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Real behavior proof

Verification

Uh oh!

clawsweeper Bot commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

amersheeny commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

amersheeny commented Jun 12, 2026

Uh oh!

clawsweeper Bot commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

amersheeny commented Jun 17, 2026

Uh oh!

clawsweeper Bot commented Jun 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

amersheeny commented Jun 17, 2026

Uh oh!

amersheeny commented Jun 18, 2026

Uh oh!

amersheeny commented Jun 18, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

amersheeny commented Jun 12, 2026 •

edited

Loading

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

clawsweeper Bot commented Jun 12, 2026 •

edited

Loading

clawsweeper Bot commented Jun 17, 2026 •

edited

Loading