Skip to content

feat: end each quiet turn on an end token the acp layer drops - #32302

Merged
KiroCrewCaptain merged 1 commit into
mainfrom
fix/19276-quiet-end-token
Nov 20, 2026
Merged

KiroCrewCaptain merged 1 commit into
mainfrom
fix/19276-quiet-end-token

Conversation

@bolichen97

@bolichen97 bolichen97 commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: A turn that ends on nothing_to_do leaves the backend's history on a real assistant message, so the backend stops inserting its placeholder and the model stops copying it, while every Kiro Crew surface behaves exactly as it does today.

Today the nothing_to_do result tells the model to write nothing. The backend's history then ends on a tool result, the backend fills the gap with its own self-describing placeholder on the next request, and the model reads it back and writes it again after later quiet ends. #19265 stopped the dashboard from showing that text, but the model still writes it: billed output on every quiet patrol cycle, one Turn-end contract violation warning per cycle, and saved copies that every later request carries (#27655 counted 190 in one backend session).

Why it matters

Every quiet patrol, heartbeat or worker-report wake pays for text nobody reads, and the copies grow the backend's history until it compacts. The warning that exists to catch real contract violations fires on every quiet cycle, so it stops meaning anything.

Not a goal

What changed (motivation → approach → change)

The maintainer decision on #19276 picks the token reply with the drop at the source: after nothing_to_do the model writes one fixed token, and Kiro Crew drops that exact token where the model's text first arrives, so no runner needs its own handling.

Part Change
Token QUIET_END_TOKEN in src/kiro_crew/session_directive.py is (quiet). The nothing_to_do result now says "Reply with exactly (quiet) and nothing else, and call nothing after this".
Drop New acp/end_token.py: drop_end_token removes a text segment (the text between two other events) whose stripped text equals the token. Chunks are held only while they are still a prefix of the token and total at most 64 characters, so a long run of whitespace is never held; anything else goes out unchanged, in order, with its message id. Every caller of the two turn loops reads them through it: stream_unsolicited (acp/client_turn.py), stream_command and _dispatch_prompt_events (acp/client_turn_control.py) for AcpClient._dispatch_events, and _run_turn (acp/session_handle_turn.py) for AcpSessionHandle._dispatch_events. So prompt, agent-started and command turns on every backend pass through it. No tool-name check and no per-backend branch.
Wording The other model-facing sites stop saying "write nothing after it": the tool description (mcp_tools/control_parts/schemas.py), the base prompt's turn-end contract (config/prompt.md), the Bot operating mode and the Bot's work-ledger patrol rule (crewmate_prompts.json), and the heartbeat prompt (dashboard/goal_reminder.py, which names the token).
Specs acp-client.md gains "Quiet-end token drop"; session.md and mcp.md describe the end token.
Prompt previews The two test/fixtures/crewmate_prompts/ goldens (read with scripts/crewmate_prompt_preview.py) change only by the Bot operating mode and heartbeat wording above.

Consumers do not change: with the token gone they see what they see today when the model writes nothing after the call. The dashboard runner records the quiet step with no late text and no contract-violation count; a refused quiet end still gets the ladder and Resume; the channel driver, cron (empty_reply_placeholder), heartbeat (_No response._) and the worker-report wake read the same empty reply as before.

The recorded turn-end decision this changes is superseded by the 2026-11-07 decision entry from #32257, as docs/decisions/README.md requires.

Backwards compatibility

Compatible: every consumer reads the same events as today unless a text segment is exactly (quiet), which is what the model is now asked to write after nothing_to_do; a backend that does not write it behaves as it does today.

Removes nothing: no key, route, field or function is removed, and any text other than the token passes through unchanged.

Tests

  • test/test_acp_end_token.py: the token alone, split across chunks and padded with whitespace is dropped; a segment padded past the 64-character hold bound is released unchanged; text that only starts like the token is released byte-identical, in order and with its message ids; each segment between other events is judged on its own (a token before a tool call is dropped and the call passes, a segment that stopped matching does not keep the next one from matching, a token prefix left at a boundary is released, a token at the end of the stream is dropped); a token-only prompt turn through AcpClient.stream_events yields no text.
  • test/test_acp_runtime_dispatch_tools.py::test_a_token_only_reply_after_a_tool_reaches_no_consumer: the same through AcpSessionHandle.prompt.
  • test/test_acp_client_agent_started_turns.py::test_an_agent_started_turn_that_is_only_the_end_token_yields_no_text: the same through AcpClient.stream_unsolicited.
  • test/test_nothing_to_do_directive.py: the tool result names the token.

The consumers need no new test: they never see the token, and their empty-reply paths (the ladder on a refused quiet end, cron's empty_reply_placeholder) are pinned by test_a_person_s_turn_cannot_end_quietly and test_cron_silent_run_placeholder.py as they stand.

Determinism: no clock and no I/O race; the loop tests use the existing fake-frame harnesses. Ran the touched and neighbouring files at -n 4 (bindings, both turn loops, the session-handle suites, directive and prompt tests), and the new tests repeated and in reversed order.

Manual verification

Still required, after merge: on an affected host (a crewmate DM on KAS), once its backend session next compacts, the Turn-end contract violation count per quiet cycle should drop to zero. That host's reporter is asked on #27655.

Real-machine verification

Machine: Linux workstation, Amazon Linux 2023 x86_64, 32 vCPU / 123 GiB, Python 3.12.14
Before (main 12902fb): python -m pytest -n0 -q -p no:randomly -p no:cacheprovider test/test_rmv_quiet_end.py (with PYTHONPATH set to an export of that commit's src), a two-test probe copied in for the run (the nothing_to_do result names (quiet); a prompt turn through AcpClient.stream_events whose only reply is (quiet) yields no text) -> 2 failed: the result says "Write nothing and call nothing after this", and the consumer receives ['(quiet)']
After (this patch on that main, tree 508bec271e89): the same command -> 2 passed
Cleanup: the probe was deleted from the checkout (git status --short empty) and the base export removed

Related Issues

Closes #19276

Decision: #19276 (comment)
Superseding decision entry: #32257
Related: #27655, #19265

Checklist

  • A Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...); PR Hygiene does not count commits
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

@bolichen97
bolichen97 requested review from a team and KiroCrewCaptain as code owners November 7, 2026 12:10
@github-actions

github-actions Bot commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ⏭️ skipped

Revision 1626437a50b1b0cb042751a6d403822574fdbc5c touches no user-facing surface (no changes under website/ or committed screenshots), so the UX review was skipped. Does not block merge.

@github-actions

github-actions Bot commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of 1626437a50b1b0cb042751a6d403822574fdbc5c — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: PASS

Nothing blocks. Every item comes from the maintainer decision recorded on base (docs/decisions/2026-11-07-a-quiet-turn-ends-with-an-end-token-kiro-crew-drops.md), and the edited test pin is the reversal that entry declares.

Not checked: whether KAS stops adding its placeholder once history ends on (quiet). The PR defers this to after merge (#27655).

What this change ships

Inventory (6 items) — 6 justified

Intent: stop the model re-writing the backend's quiet-end placeholder (billed text and false contract-violation warnings) while every surface looks as it does today. ADDITION, implementing a recorded decision.

  1. After nothing_to_do, the tool result tells the model to reply with exactly (quiet) instead of nothing — justified
  2. Any text segment that is exactly (quiet) is dropped before any consumer sees it, on prompt, agent-started and command turns — justified
  3. The base prompt, the Bot operating mode, the patrol rule and the nothing_to_do description now say "write only the end token" — justified
  4. The heartbeat prompt names (quiet) literally and tells the model nobody sees it — justified
  5. The test pin "Write nothing and call nothing" is rewritten to the token wording; the description declares this as superseding the 2026-10-03 entry — justified
  6. Specs: acp-client gains "Quiet-end token drop"; the explanation of where the placeholder comes from moves from session.md's late-text paragraph into its new end-token paragraph — justified

[FIRST-PRINCIPLES-REVIEWED] 1626437

Description read

[DESCRIPTION-READ] 0c2592a967db116d12d4bf40be4452639c337f44a8daaccba3257f038341ec7d

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@bolichen97
bolichen97 enabled auto-merge (squash) November 7, 2026 12:14
@github-actions github-actions Bot added the readiness: action required A blocking check or review needs attention label Nov 7, 2026
@github-actions

github-actions Bot commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of 1626437a50b1b0cb042751a6d403822574fdbc5c and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 1626437

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 1626437a50b1b0cb042751a6d403822574fdbc5c false-positive: <reason, 20 to 500 characters>

@github-actions github-actions Bot added the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Nov 7, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

self-added: no

  • Side-channel events release fragments of a quiet-end token (span=b654f0c8ee98)

Rebutted: disproportionate; the effect is one operator-log count, never text a person sees.
Text after an applied terminal directive is dropped at arrival and only counted on both consumers: dashboard/chat_turn/stream_chunks.py (the _terminal_directive_applied branch returns None) and messaging/driver.py (text_after_terminal). Released fragments are never streamed, persisted or posted, and the turn still ends quiet.
The trigger needs a non-text event to land between two chunks of a short token after the quiet end; the worst case is one Turn-end contract violation warning line.
Matching across "side-channel" events needs a list of event kinds that every future kind must be sorted into, which costs more than that log line; the adjudication on this head downgraded it to LOW for the same reason.
No code change.

[operator: bolichen97]

@bolichen97

Copy link
Copy Markdown
Contributor Author

self-added: no

  • The terminal contract reverses an unsuperseded recorded decision (span=01d592f547c0)

Intended: this PR acts on the maintainer decision recorded at #19276 (comment) .
The superseding entry ships in its own pull request, #32257, as docs/decisions/README.md rule 3 requires; the 2026-10-03 entry stays byte-identical there.
This is the finding's own fix: once #32257 merges, this branch rebases onto it and the recorded decision is superseded in the tree.
No code change here; the block clears when #32257 lands, or through a maintainer owner-decision override that links that decision.

[operator: bolichen97]

@bolichen97
bolichen97 force-pushed the fix/19276-quiet-end-token branch from 70063ec to b2b4f07 Compare November 7, 2026 17:34
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in test/fixtures/crewmate_prompts/ledger.golden.txt (a conflict, regenerated with scripts/crewmate_prompt_preview.py --fixture ledger --write), docs/architecture/mcp.md, docs/system-specs/modules/acp-client.md and docs/system-specs/modules/session.md since this head's CI ran (bab2387 -> 6f786d9); CI re-running.

The new head b2b4f07 also clears the Flake Pattern Ratchet red: the two added tests pass FALLBACK_TIMEOUT_SECS from kiro_crew.testing.wait instead of a literal timeout (K19).

[operator: bolichen97]

@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge labels Nov 7, 2026
@github-actions

github-actions Bot commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of 1626437a50b1b0cb042751a6d403822574fdbc5c — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

No design-level concerns: one drop point that every caller of both turn loops goes through, as the 2026-11-07 decision entry asks for, with no per-backend branch and no hunk the description doesn't account for.

Notes

[DESIGN-REVIEWED] 1626437

Description read

[DESCRIPTION-READ] b9c2a2ae2392c17e505efc95548fd4a9f63cd1e742ff7f72fe639536d62ca72d

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed 1626437a50b1b0cb042751a6d403822574fdbc5c — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 1626437

Verdict parsed from the review's SHA-scoped output markers for commit 1626437a50b1b0cb042751a6d403822574fdbc5c.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 1626437a50b1b0cb042751a6d403822574fdbc5c false-positive: <reason, 20 to 500 characters>

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge and removed readiness: checking Automated validation is still running labels Nov 7, 2026
@bolichen97
bolichen97 force-pushed the fix/19276-quiet-end-token branch from b2b4f07 to a4ad1af Compare November 7, 2026 20:39
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in docs/system-specs/modules/session.md (a conflict: the quiet-end paragraph keeps main's held-text-for-a-pending-steer sentence from #32396 and this PR's end-token paragraph), src/kiro_crew/acp/client_turn.py, src/kiro_crew/acp/client_turn_control.py, test/test_nothing_to_do_directive.py, docs/architecture/mcp.md and docs/system-specs/modules/acp-client.md since this head's CI ran (6f786d9 -> 03016b6); CI re-running.

The new head a4ad1af also clears the Backend Tests shard 5 red (Linux and Windows): this PR rewords conductor.how_you_work, so test/test_context_composition_contract.py now pins the new digests of member_how_you_work and member_how_you_work_unavailable. Every _dispatch_events caller on the new base still goes through drop_end_token.

[operator: bolichen97]

@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge readiness: checking Automated validation is still running labels Nov 7, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

self-added: no

  • The turn-end contract changes without a superseding decision entry (span=eb7326b784fc)

Rebutted: this PR acts on the maintainer decision recorded on the issue: #19276 (comment) .
The superseding docs/decisions/ entry, with its Supersedes: line, ships in its own pull request, #32257, as docs/decisions/README.md rule 3 requires; the 2026-10-03 entry stays byte-identical there.
Once #32257 merges, this branch rebases onto it and the tree carries that entry, which is this finding's own fix.
No code change here: the block clears when #32257 lands, or through a maintainer owner-decision override that links that decision.

[operator: bolichen97]

@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in src/kiro_crew/acp/session_handle_turn.py, docs/system-specs/modules/session.md and test/fixtures/crewmate_prompts/ledger.golden.txt since this head went green (df73916 -> 2f8704c); CI re-running.

GitHub reported the previous head as conflicting, which is the one case this branch rebases for. Resolutions, keeping both sides:

Every added and removed line is byte-identical to the previous head; only context moved. Main still has the same four _dispatch_events callers and all four are wrapped.

Retrospective on this head (the rounds view called for one): 13 mechanisms ruled, all kept, none removed, narrowed or reverted. Every review lane on the previous head was green; it only conflicted.

resolution: kept -- this is the 4th push after three blocked heads, so the approach was re-examined before pushing again. No span has been fixed twice and no line-level finding is open: the three blocked heads were the merge-order block that waits on #32257, main's own red tests, and one run of CI API rate-limit failures, none of them a finding in this change. The code is therefore unchanged from the head that went green, and the only edit is the re-sync above.

[operator: bolichen97]

@github-actions github-actions Bot removed the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Nov 13, 2026

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approving at the maintainer's request.

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: passed Eligible automated validation passed for the current revision labels Nov 13, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

Re-sync budget reset

The GPT 6.1 lane found a real defect on 3ff5f6c: the quiet-end token hold in acp/end_token.py had no size bound, so a long run of whitespace-only text chunks was held and re-joined on every arrival. The fix needs one push, and a push has to sit on current main, which has moved since this head went green. This PR is inside its 48-hour watch window, so the re-sync count restarts here and the fix goes out on a fresh main.

[operator: bolichen97]

@github-actions github-actions Bot added the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Nov 15, 2026
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Nov 19, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

self-added: no
mechanism: the quiet-end prefix hold in drop_end_token is bounded to _HELD_CHARS_MAX raw characters
resolution: narrowed -- the prefix hold is bounded to _HELD_CHARS_MAX raw characters; overflow takes the existing release path, so no flag or new state

  • Whitespace leaves the token buffer unbounded (span=b654f0c8ee98)

fixed in 0950e7a: a segment whose held raw text passes _HELD_CHARS_MAX (64) characters is released unchanged and stops matching until the next event, so the hold and the work per arrival are both bounded.
Both dispatchers yield only non-empty text chunks (if chunk: in acp/client_turn.py, if text: in the shared session-update parser), so 64 characters also caps the held chunk count: one bound covers both, with no count cap or overflow flag.
The token is 7 characters; a real quiet end (the token plus a newline or two) stays far under the bound and is still dropped.
Pinned by test_a_segment_longer_than_the_hold_bound_is_released_unchanged (red on the old hold, green now); 32,000 chunks of four newlines took about 31 s through the old hold and about 0.01 s now.

[operator: bolichen97]

@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in docs/architecture/mcp.md, docs/system-specs/modules/acp-client.md, docs/system-specs/modules/session.md, src/kiro_crew/acp/client_turn.py, src/kiro_crew/acp/client_turn_control.py, src/kiro_crew/acp/session_handle_turn.py, src/kiro_crew/config/prompt.md, src/kiro_crew/crewmate_prompts.json, src/kiro_crew/mcp_tools/control_parts/directive.py, src/kiro_crew/mcp_tools/control_parts/schemas.py, src/kiro_crew/session_directive.py, both test/fixtures/crewmate_prompts/ goldens, test/test_acp_client_agent_started_turns.py, test/test_context_composition_contract.py and test/test_nothing_to_do_directive.py since this head went green (2f8704c -> 017551d); CI re-running.

This rebase onto current main clears the merge conflict GitHub reported. Main reworded the Bot operating-mode text around the changed nothing_to_do clause and moved the two member_how_you_work* hash pins; main's lines are kept, the clause is unchanged, and both pins are recomputed from the merged text. In session.md, main's "The placeholder at a turn's start." and this change's "The end token." are both kept, this one first.

What was fixed: the GPT 6.1 finding on 3ff5f6c (the quiet-end token hold could grow without bound on whitespace). The hold is now capped at 64 characters, after which the text is released unchanged; its disposition is the comment above. Every other added and removed line is unchanged.

Main's other edits sit beside this change: the four _dispatch_events callers are still the only ones and all four are wrapped. On this same patch one step earlier on main (main's later commits touch no source file) the lint, import-order, formatting and type checks, the repository's static-analysis rules, the docs checks and 3,409 related tests (both turn loops, the shared-runtime dispatcher and watchdogs, agent-started turns, the session-handle bindings, the channel driver, the prompt hash pins and goldens) passed, and the two-test before/after probe in the PR body fails on main and passes on this head.

[operator: bolichen97]

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge and removed merge conflict Branch has merge conflicts with its base — author must resolve before merge readiness: checking Automated validation is still running labels Nov 19, 2026
After nothing_to_do the model now replies with one fixed token, so the
backend's history ends on an assistant message instead of a tool result
and the backend stops inserting the placeholder the model kept copying.
Kiro Crew drops that exact token where the model's text first arrives
in both turn loops, so every surface sees the same empty reply as today.
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in docs/architecture/mcp.md, docs/system-specs/modules/acp-client.md, docs/system-specs/modules/session.md, src/kiro_crew/acp/client_turn.py, src/kiro_crew/acp/session_handle_turn.py, src/kiro_crew/config/prompt.md, src/kiro_crew/crewmate_prompts.json and test/test_context_composition_contract.py since this head went green (017551d -> faec552); CI re-running.

This rebase onto current main clears the merge conflict GitHub reported. The two member_how_you_work* hash pins conflicted because #39413 moved them on main; both are recomputed from the merged text.

What changed besides the rebase: #39413's new Bot patrol rule ended "Then write no text at all and end with nothing_to_do.", which contradicts this change's rule that after nothing_to_do the model writes only the end token. That sentence now goes on "After it, write only the end token its result gives you.", the same wording as the Bot operating mode. Its original words are kept, so test_mate_conductor_parity.py still holds, and no golden moves. Every other added and removed line is unchanged.

Main's other edits to these files sit beside this change: apart from #39413, none of them names nothing_to_do. On this patch on current main: the related backend test set (28,987 + 8,347 passed), the formatting, import-order, lint and type checks, the scripted backend gates and docs lint, the repository's static-analysis rules and the flake ratchet passed, and two independent local reviews found nothing. The two-test before/after probe in the PR body fails on main and passes on this patch.

[operator: bolichen97]

@bolichen97
bolichen97 force-pushed the fix/19276-quiet-end-token branch from 0950e7a to 1626437 Compare November 20, 2026 02:09
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: passed Eligible automated validation passed for the current revision and removed readiness: action required A blocking check or review needs attention merge conflict Branch has merge conflicts with its base — author must resolve before merge readiness: checking Automated validation is still running labels Nov 20, 2026

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I reviewed head 1626437 and checked that all four callers of the two ACP turn loops read their events through the new quiet-end token drop. The filter only holds text that could still turn into the token, caps the hold at 64 characters, and releases everything else unchanged, in order and with its message ids. On this head the new tests, the prompt goldens and hash pins, and the wider ACP client, runtime and session handle suites all pass. Removing the wraps makes the three new integration tests fail, so they test what they claim. I found no wrong behaviour, crash, security problem or data loss, and the superseding decision entry is already on main, so nothing blocks.

@KiroCrewCaptain
KiroCrewCaptain merged commit 1c153aa into main Nov 20, 2026
99 checks passed
@KiroCrewCaptain
KiroCrewCaptain deleted the fix/19276-quiet-end-token branch November 20, 2026 03:22
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Nov 20, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

feat: end a quiet turn on a valid assistant turn so the backend needs no alternation filler

2 participants