The isTerminal:false early-return skipped #finishAgentEnd, the only site
that flushes a deferred plan-mode model switch, so an automatic continuation
(async wake) ran on the old model/thinking level until the terminal settle.
Flush the pending switch on the non-terminal branch before returning; the
title/loader teardown stays deferred to the terminal agent_end.
EventController.#handleAgentEnd guarded only on session.isStreaming, so a
non-terminal agent_end (isTerminal:false, emitted while an async job will
re-wake the loop) flipped the terminal title to idle and tore down the
working loader while a /vibe worker or async bash job was still running.
Early-return on event.isTerminal === false, matching the guard every other
agent_end consumer already applies; the later terminal agent_end performs
the normal teardown.
Fixes#7386
Address Codex review on #6881. Retraction of never-run tool cards no longer runs eagerly at message_end: only a TTSR rewind (known via isTtsrAbortPending) retracts there. A terminal error/abort lets agent-loop's synthetic tool_execution_end settle each card in place so the failure stays visible, and an auto-retry removes those synthetic-settled cards only when auto_retry_start actually supersedes the turn.
Also settle a held server-resolved (Cursor/todo) completion after a mid-stream tool-call id re-key, so the migrated card is not stranded pending.
Fixes#6879
- Implemented clipboard register management, parsing, and execution rules for CUT, COPY, and PASTE operations in the hashline engine.
- Added session-persistent clipboard state and integration across agent session execution, diff previews, and streaming tools.
- Added comprehensive validation, error messages, recovery handling, and test coverage for clipboard and block operations.
handleEvent has no blanket pre-render (removed for issue #4353), so
each handler must explicitly schedule one when it changes something
visible -- every other statusLine.invalidate() call site in this file
pairs it with ui.requestRender(). The new model_changed handler only
invalidated the cache, so an internal model switch (prewalk hand-off,
retry-fallback) while the TUI was otherwise idle left the status line
showing the stale model until an unrelated event happened to render.
Flagged by @chatgpt-codex-connector on PR #6908.
(cherry picked from commit f565e3a4956bda467b6679a3c3f1e48379395a78)
AgentSession#setModelWithProviderSessionReset is the single choke point
every model mutation runs through (explicit /model, prewalk hand-offs,
retry-fallback, model cycling). It previously changed agent.state.model
silently — no session event told subscribers (ACP, RPC, TUI) that the
active model moved.
Emit a new model_changed AgentSessionEvent from that choke point
whenever the model actually changes, and wire it into every consumer
that must exhaustively handle AgentSessionEvent: the TUI event
controller (invalidates the status line, same as thinking_level_changed)
and the RPC client's forwarded-event allowlist.
(cherry picked from commit f76325de2c7821dd7046ddb67546577c3575a263)
GitHub Copilot's call_id|id transport (and any stream that delivers a tool's name/args before its id) makes a streamed tool-call block appear with an empty or partial id, then rewrites it in a later delta. The transcript keyed the live card by that mutable content.id, so the changed id passed the creation guard and spawned a second card: the old-id card orphaned as a blue pending preview while the new-id card took the result.
Track the streamed id per tool-call block position (reset per assistant message) and migrate the live card's id-keyed trackers (pendingTools, tool timeline, args reveal, read tracking, and the shared read group's entry) when the id changes, so the populated id reuses the existing card.
Fixes#6879
When a successful read result was persisted before its live tool_execution_end reached the UI, a transcript rebuild replayed the completed card and removed the pending handle. The delayed read completion then created a fallback ReadToolGroupComponent beside replay.
Treat the existing tool timeline entry as proof that the read already had a card; clear stale read tracking and skip fallback creation. Added an exact memory:// success -> persisted replay -> delayed completion regression.
Fixes#6879
Track tool-call ids whose failed-attempt cards were retracted until agent-loop emits their synthetic skipped-tool completion. Ignore the synthetic start and consume the matching end so the no-pending handler cannot recreate the failed card.
Extended the retry regression through the exact message_end -> synthetic tool start/end -> auto-retry sequence.
Fixes#6879
Seal an uncommitted pending tool card before removing it from the transcript so ToolExecutionComponent cancels its spinner, todo strike, and edit-preview work instead of continuing to schedule renders while detached.
Added a fake-timer regression proving an animated write card renders while pending and stops component-scoped renders after an aborted turn retracts it.
Fixes#6879
Reset #lastReadGroup before retracting a superseded grouped read so the retry creates and mounts a fresh group instead of updating the detached component. Clear the removed call's read tracking as well.
Added regression coverage for an aborted grouped read followed by a visible successful retry.
Fixes#6879
An assistant turn that streamed tool calls and then ended with error/aborted
(a provider blip that auto-retries, a TTSR rewind) has its never-run cards
dropped from the active context, but EventController#handleMessageEnd only
sealed them in place. The retry re-streamed fresh cards below, so each call
rendered twice.
Retract the never-run pending tool cards still in the live region (never
committed to native scrollback, so removable) and forget them from pendingTools
and the timeline map at message_end; a card already on the scrollback tape is
still sealed in place. Follow-up to #6516 / #6519.
Fixes#6879
When Cursor packs toolCallStarted and toolCallCompleted into one HTTP/2
chunk, the bridge tool_execution_end (synchronous callback, fired
mid-parse) reaches the interactive controller before the streamed
toolcall_start (queued on AssistantMessageEventStream, delivered a
microtask later). The controller found no pendingTools entry, dropped
the completion, and the card created afterwards animated forever.
Two halves, each necessary:
- Hold an early todo completion in #orphanedToolCompletions and replay
it when the streamed block creates its component.
- Guard card creation from cumulative message_update frames with the
turn-scoped #toolTimelineComponents map. Without this, the update
after the replay re-lists the same toolCall block, finds pendingTools
empty again, and spawns a second, permanently pending card. This is
also why emitting a synthetic tool_execution_start from the bridge
(previous attempt, reverted) could not work.
Both maps are cleared together at the existing transcript-anchor reset
sites. The normal ordering (start first) is covered by a control test.
Two review findings on the native todo sync.
- TodoItem.dependencies is a graph the local model cannot store: rows
are keyed by content, carry no id, and hold no edges. An imported
dependent row files as plain pending and nextActionableTask then
offers work the server considers blocked. Refuse snapshots with an
edge pointing at an unfinished row; edges whose blockers already
finished constrain nothing and still mirror.
- The todo failure warning interpolated the provider error verbatim.
Collapse and truncate it at the render boundary.
Also documents two known, unfixed defects: an async cursorOnToolResult
transformer resolving after the buffer drain, and the todo card
lifecycle race. Emitting a synthetic tool_execution_start for the
latter was measured and rejected -- the completion deletes the entry it
creates, so the late streamed block adds a second card.
Moved vocalizer interruption from agent_start to user message_start so agentLoopContinue preserves queued speech.
Expanded regression coverage for continuation starts and user prompt boundaries.
Fixes#6375
Moved vocalizer clearing from each assistant/tool continuation boundary to the start of a new agent run.
Added regression coverage for uninterrupted internal turns and new-run interruption.
Fixes#6375
Kept the bare 'π' brand per owner direction: the separator between the
brand and the session label now carries the state — '>' when it's the
user's turn (idle), animated spinner frames while working, '!' when the
agent is blocked on the user. Disabled ('tui.titleState' off) renders the
pre-state 'π: label' layout. Updated the state/runtime tests to the new
contract.
Preserved settled user and assistant component instances during compaction-only transcript rebuilds so warmed Markdown and layout caches remain valid.
Covered both manual and automatic compaction paths with focused regression tests.
Fixes#6033
- Added the `xd://` virtual device protocol (`internal-urls/xd-protocol.ts`, `tools/xdev.ts`): tools declaring `loadMode: "discoverable"` are unmounted from the request tools array and driven via `read xd://` (list/docs+schema) and `write xd://<tool>` (execute), gated by the `tools.xdev` setting (default on) and inlined into the system prompt.
- Merged the `irc`, `job`, and `launch` tools into a single `hub` tool (`tools/hub/`, `async/job-manager.ts`): messaging keeps `send`/`inbox`/`list`, job control maps to `wait`/`cancel`/`jobs`, process supervision keeps `start`/`logs`/`stop`/`restart`/`describe` with `ps`, and the unified `wait` races background jobs against peer messages; SDK `IrcTool`/`JobTool`/`LaunchTool` are replaced by `HubTool`.
- Removed the hidden `resolve` tool in favor of the `xd://resolve`/`xd://reject`/`xd://propose` resolution devices, auto-including `write` whenever a deferrable tool or plan mode is present.
- Removed the BM25 tool-discovery system: the `search_tool_bm25` tool, the `tool-discovery` module, the `tools.discoveryMode`/`mcp.discoveryMode`/`mcp.discoveryDefaultServers`/`tools.essentialOverride` settings, per-tool MCP selection, and the `mcp_tool_selection` message type.
- Unified tool presentation on `ToolLoadMode` (`essential`|`discoverable`), replacing the custom-tool `xdev?: boolean` opt-out; custom, extension, MCP, RPC host, image-generation, and TTS tools now default to `discoverable`, and added a `satisfies` predicate to `SoftToolRequirement`.
- Removed the standalone `ssh` command tool and `ssh/ssh-executor` (the `ssh://` read/write/search protocol stays), and made `--tools` address hidden built-ins.
- Updated collab-web to render `xd://` dispatches and `hub` op families, dropped the `search_tool_bm25`/`ssh`/`report-finding` renderers, refreshed tool docs and prompts, and migrated the affected tests and changelogs.
- Introduced conditional scrollback clearing during UI renders when transcript compaction is enabled.
- Updated `CommandController` and `EventController` to respect the `display.collapseCompacted` setting.
- Configured `SelectorController` to trigger a chat rebuild and UI reset when the compaction setting changes.
- Updated `InteractiveMode` to dynamically toggle between collapsed and full inline history based on user settings.
- Stop propagating real-time updates for backgrounded Bash jobs to avoid UI flickering once a job enters the background.
- Refine background task tracking in `EventController` to distinguish between persistent background tasks and transient backgrounded Bash commands.
- Update UI rendering to display cleaner background job metadata in the footer instead of inline text notices.
AgentSession defers and coalesces the wire-level agent_end while a
prompt is in flight (#emitSessionEvent), so a multi-attempt retry saga
often surfaces only ONE agent_end to EventController — which can be
the final settle, not an intermediate attempt. Consuming #retryPending
against whichever agent_end arrived first (previous commit) could
therefore discard the real final failure notification.
Switch to gating purely on the retry lifecycle: #retryPending is set
by auto_retry_start and cleared only by auto_retry_end (both
outcomes), never consumed by sendErrorNotification itself. Those
lifecycle events are never deferred, so they reliably bracket the
window a retry is actually outstanding regardless of how agent_end
coalescing lands.
Close the residual gap this creates: #handleRetryableError's
classifier-refusal and Fireworks-fallback-ineligible branches could
short-circuit a saga that already announced auto_retry_start without
ever emitting auto_retry_end, latching #retryPending open forever.
Both branches now emit a final auto_retry_end(false) when a prior
attempt already started the saga. #handleAgentStart also clears
#retryPending defensively so a saga that still somehow never resolves
cannot suppress a later, unrelated turn's notification.
A retryable error's agent_end fires with the failed assistant message
(stopReason === 'error') the instant #handleRetryableError schedules a
retry (auto_retry_start), before the retry has a chance to recover.
sendErrorNotification read that transient agent_end the same as a real
final settle, so error.notify=on raised a 'Stopped with error' toast
even for turns that went on to succeed on retry.
Track a #retryPending flag set on auto_retry_start and consumed
(check-then-clear) by sendErrorNotification, so exactly one mid-retry
agent_end is suppressed per attempt. #handleAutoRetryEnd also clears it
directly on both success and failure so a recovered retry never leaves
it stuck; consuming it on read (rather than only via auto_retry_end)
keeps it self-healing for the classifier-refusal short-circuit path,
which can return from #handleRetryableError without ever emitting a
fresh auto_retry_end.
Daniel Anderson-Little