- Added `#hasPendingAsyncWake` to detect running or pending background jobs owned by the agent.
- Deferred todo reminders and `session_stop` hook passes until all agent-owned background async jobs complete.
- Ensured scheduling pauses caused by async jobs do not trigger terminal session stops or premature todo nags.
Cursor's provider only pushed toolCall content blocks for MCP and todo
in processInteractionUpdate.toolCallStarted. Native tools (bash, read,
write, grep, ls, delete, lsp) execute via the exec channel and produced
no toolCall blocks, so persisted assistant messages contained only text.
On replay, renderSessionContext could not pair the subsequent toolResult
messages with any toolCall block and fell through to addMessageToChat
(a no-op for toolResult), causing header-less \`\\u23ce\` output beneath the
last assistant text.
- packages/ai/src/providers/cursor.ts: add synthesizeCursorExecToolCall
and inject it at the top of each native exec case in
handleExecServerMessage, using the coding-agent bridge's mapped tool
name and args so live event and rebuild render identically. Normalize
args.toolCallId before invoking the handler so provider block id and
bridge result id always match.
- packages/agent/src/agent.ts: drop the text-length split in
#emitCursorSplitAssistantMessage. With toolCall blocks now at their
correct positions in content, emit the assistant message as-is
followed by buffered toolResults; the split's preambleText-per-text
copy also silently duplicated text on multi-block turns.
- packages/ai/test/cursor-streaming-args.test.ts + new
packages/coding-agent/test/issue-4348-repro.test.ts: guard block
ordering, event sequence, and rebuild pairing behavior.
Fixes#4348
- Simplified match logic to rely exclusively on content hash equality.
- Removed strict validation that rejected colliding snapshot tags.
- Updated recovery behavior to resolve collisions to the most-recently recorded snapshot.
- Refactored tests to expect successful preview and patching despite tag ambiguity.
- Added `getModel` to `AgentLoopConfig` to allow runtime model resolution.
- Updated `streamAssistantResponse` to resolve the model dynamically per provider call instead of using the stale configuration snapshot.
- Enabled mid-run model switches to take effect immediately for context promotion and retry fallbacks.
Anthropic's OAuth usage endpoint now ships a generic limits[] array;
model-scoped weekly caps (Fable) exist only there while the legacy
seven_day_opus/seven_day_sonnet buckets are permanently null. Parse
weekly_scoped entries into anthropic:7d:<slug> tier rows (Claude 7 Day
(Fable)), backfill shared 5h/7d from session/weekly_all when legacy
buckets are absent, and accept limits[]-only payloads in hasUsageData.
Add scopeLimits/blockScope to claudeRankingStrategy so an exhausted
Fable/Mythos cap gates only matching-model requests instead of cooling
down the whole OAuth credential (mirrors the Antigravity per-counter
precedent). Dedupe the ACP /usage tier suffix when the label already
names the tier.
- Changed the `path` property from an array of strings to a single semicolon-delimited string across tool definitions and tests.
- Updated validation error messages to reflect the new `path` input format.
- Adjusted all relevant test cases to provide path targets as semicolon-separated strings.
- Restored released sections byte-identical to pre-sweep main (immutable).
- Consolidated the 18 merged-fix entries under [Unreleased].
- Dropped a foreign entry imported from a non-landed branch (edit-tool
oldText/newText snapshot bounding), which this merge set does not contain.
Plan mode records idle IRC deliveries into context without waking a turn,
so an 'irc send await:true' sender was stranded until its wait timeout:
delivery reported injected but no reply could ever be generated. Extend the
existing ephemeral side-channel auto-reply (previously only for mid-turn
recipients with async execution disabled) to the idle plan-mode case — the
second situation where a real reply turn cannot happen in time. No primary
turn is woken; plan-mode convergence stays user-driven.
The deliverIrcMessage eligibility change itself rode in the previous commit
(same-file hunks); this commit carries the bus doc and the regression test:
an awaited idle IRC message in plan mode resolves the sender's bus waiter
via the side-channel reply while the primary loop stays asleep.
The settle-time reminder queues a hard 'required' tool choice paired with a
scheduled continuation. If that continuation never runs (user prompt bumps
the generation, dispose, compaction/handoff) or plan mode is exited first,
the queued directive leaked onto the next unrelated turn as a forced tool
call. Remove it by label on continuation skip, on user-initiated prompts,
and when plan mode is disabled.
Main removed the floating todoReminderContainer in 112317bc8 (todo
reminders are now anchored inside the scrollback transcript and reset
by renderInitialMessages({clearTerminalHistory: true})), so the added
this.todoReminderContainer.clear() references a property that no longer
exists post-merge, failing typecheck and throwing on every /btw branch.
Revert the interactive-mode hunk and its test, and reword the changelog
entry to cover only the goal-mode todo context fixes.
The github renderer materializes plain Text per display rebuild, so its
animatedPendingPreview opt-in requested 30fps repaints with a frozen
glyph for the whole run_watch wait; drop the flag. Custom tools with
only one of renderCall/renderResult never route to the animated generic
fallback, so gate the unregistered-renderer spinner on both being
absent. Regression tests for both no-tick contracts.
The consume-once source map kept entries for graph modules the initial
import never loaded (modules only reached via lazy dynamic imports).
Their first import - possibly long after load, and after an on-disk
edit - was served the boot-time snapshot instead of current file
content, and the unconsumed sources stayed in the plugin closure for
the process lifetime.
Clear the map once the entry import settles: everything Bun loaded at
startup was already consumed (keeping the read-once win), and anything
left must be read at its actual import time, matching pre-dedup
behavior for lazy modules. The new regression test passes on the
pre-dedup baseline and fails on the unfixed dedup.
isAllCapsWord matched any multi-letter token without a lowercase letter,
so CJK tokens registered as ALL-CAPS words: two adjacent ones marked the
whole source shouty and silently disabled acronym restoration for every
non-Latin-script message (e.g. '修复 CNPG 集群故障' kept 'Cnpg'). Require an
actual uppercase letter; cased-script shout detection is unchanged.
The PR's changelog, doc comment, and prompt examples all name ETL as a
restored acronym, but the review-response narrowing (vowel heuristic +
allowlist) silently dropped it: ETL bears a vowel and was not listed.
Add it to COMMON_TITLE_ACRONYMS and pin it in the allowlist test.
The three-way merge against main silently dropped the entry because
main rewrote the old Unreleased region into released sections; the
branch changelog now mirrors main's head with the entry inserted under
the current Unreleased section so the merge keeps it.
Replaces the bespoke batch-scoped process.exit interceptor with the
withExitGuard convention main established for extension/hook/plugin
loaders (500c39aa2): guard the module import and factory invocation so
a synchronous process.exit()/process.reallyExit() from a custom tool
becomes an ExtensionExitError handled as a recoverable load error,
while host exit paths stay untouched outside the guarded windows.
Tests cover the import-time exit (issue #1704 repro) and factory-time
exit; both would kill the test process without the guard.
main independently absorbed batch-1 (incremental grapheme slice 718c7cea2, markdown
stream-prefix cache 705426548 + 3822a83b4) and the pathTo/patch.ts items; restore
main's refined versions wholesale. Port the model-resolver optimization onto main's
resolver shape: hoist per-candidate case folds in matchModel and build the
preference context once per role resolution (matchPatternWithContext) instead of
per fallback pattern. Rewrite changelog entries to the surviving items only.
Byte-identical copy of the withExitGuard/ExtensionExitError block from
main (500c39aa2) so the custom-tool loader can reuse the established
guard convention; merges as an identical change against main.
Narrow acronym restoration so plain all-caps English words such as FIX
and WORK do not get restored when the model naturally capitalizes the
first title word. Restorable all-caps source tokens now need a stronger
acronym signal: a common technical acronym allowlist, digits, or a
consonant-only shape.
This keeps CNPG, ETL, JWT, SQL, and API restoration while preserving the
anti-shout behavior for single emphatic words.
Fixes#4220
- Added an enhanced speech pipeline that utilizes small models to rewrite text for natural language synthesis.
- Implemented `SpeechEnhancer` and `BlockAccumulator` to manage fence-aware text streaming and paragraph splitting.
- Configured a new `speech.enhanced` setting to toggle between mechanical and enhanced vocalization modes.
- Resolved `EPIPE` rejections during speech playback by ensuring stream flushes and suppressing stop-related errors.
reconcileTitleCasing now maps ALL-CAPS source tokens (CNPG, API, ETL,
JWT) into an acronyms table and restores them when the model produces a
plain title-cased artifact (Cnpg). Restoration is disabled when the
source is shouty (>=2 consecutive multi-letter ALL-CAPS tokens like
"FIX the BUG NOW" or "ALL ERROR HANDLING"), and lowercase model output
is left alone so isolated single-word emphasis (WORK -> work) is never
re-shouted.
The three title system prompts (title-system.md, title-system-marker.md,
tiny-title-system.md) also gained an explicit instruction to preserve
ALL-CAPS acronyms verbatim, so a competent model short-circuits via the
verbatim set before post-processing kicks in.
Fixes#4220
Removed the live-context guard that let default model selection persist a new role without changing the active session model. The next prompt's compaction path now owns oversized-context recovery after a switch.
Fixes#4219
- Added `SpeakableStream` to strip markdown noise, silence code blocks and tables, normalize links and paths, and emit sentence/clause segments.
- Reworked `Vocalizer` to segment assistant deltas in the parent process, lazily open TTS streams, idle-flush partial thoughts, and chain playback sessions.
- Added gapless streaming playback with ffmpeg/sox backends, ducking-aware pacing, fallback file playback, and immediate stop handling.
- Added IPC `sendAndFlush` support and used it in the TTS worker so audio chunks drain before blocking ONNX inference resumes.
- Added speakable-stream coverage for markdown filtering, segmentation latency, idle flushing, and forced long-segment splits.
- Replaced `grep`, `glob`, and `ast_grep` `paths` inputs with optional single `path` strings while preserving default workspace-root behavior.
- Added shared `toPathList` normalization for legacy arrays and JSON-encoded arrays across tool execution and TUI renderers.
- Updated prompts, fixtures, shims, transcript summaries, and tests to send and display the new `path` argument.
- Updated collab-web search tool cards to read `path` while falling back to legacy `paths` for historical transcripts.
- Recorded the contiguous coding-agent changelog run for the tool-path breaking change and adjacent TTS entries.
can1357