The agent manager runs option, model, thinking and mode changes in the agent's
lifecycle lane; providers do no sequencing. The hook deadlock warning now covers
agent.create.before and option writes.
The create-agent preferences convert the legacy key on read, keyed on the
built-in codex provider id, so the Codex draft shows the Fast selection the user
last used. Tagged COMPAT(featureValues) with the other legacy-option shims.
The profile form always sends options. The daemon reads a write without them as
one from an app that predates options and folds featureValues over the stored
ones; a write that carries them, empty included, replaces them. A profile left
with no options is stored without the key, so removing the last one persists.
listControls now answers { controls, options } like AgentManager.listDraftControls:
the options a session opened from the draft would run with, with config
defaults merged in and anything the provider resolves natively for the ones
nobody set. ACP fills unset feature selects from the probe session's
currentValue, so a fresh Cursor draft shows the Fast selection the agent opens
with instead of no selection. The legacy features response and the MCP draft
listing already project from the manager's result, so they follow.
A stored agent with no resumable session is now reopened by
AgentManager.createAgentFromRecord, which rereads the record inside the agent's
lane before creating the first session. A metadata or label write queued ahead
is in the record it reads; one queued behind applies to the live agent. The
loader no longer passes a record it read outside the lane. createAgent itself
also runs in the lane, like resume.
The agent manager owns all sequencing for one agent: applyOptions, setModel,
setThinkingOption and setMode now run in the same lane as open, close, archive,
labels and metadata, so a provider session never sees two of them at once.
Provider-side serialization (the applying chains in the ACP, plugin and OpenCode
sessions, and Claude's apply-over-current-options handling) is deleted.
* fix(server): keep earlier replies visible after agent resume
Idle runtime collection retains timeline rows but recreates provider sessions. Scope turn counters to a fresh UUID namespace so resumed sessions cannot reuse persisted turn identities.
* test(server): require a turn ID before inspecting its namespace
* fix(app): keep show more and plan card triggers in place on expand
Long user message previews and plan cards toggled with a plain setState,
skipping the timeline's disclosure hold. On Android the expanded content
painted over the next row for a frame before the list learned the new
size and corrected the scroll. Route both toggles through
useDisclosureHold like the other timeline disclosures.
* fix(app): keep show more at the same y by holding the message's bottom edge
Show more sits under the text it opens, so a top-anchored hold pushed it off
screen. The disclosure hold takes a trigger edge; MessagePreview passes the
bottom edge. Web pins that edge in the reading anchor, native anchors the row
below the message, and while following the end pin keeps the bottom in place.
The LegendList patch counts the row just past the viewport as in view and drops
the Android clamp prediction that moved a collapse near the end to the end.
* fix(app): keep show more in place on an unanswered final prompt
The last row has no row below it to anchor its bottom on, so a show more
on a prompt the agent has not answered yet moved the trigger off screen.
The hold now mounts the end pin while detached and toggles a frame later,
once the pin anchors, and re-sticks only if the list was at its end when
the toggle was tapped.
The disclosure browser test now renders the real Markdown renderer.
* fix(app): resume following when a final prompt collapse reaches the end
An end hold decided its release from whether the list was at its end
when tapped. A collapse that leaves less scroll room than that distance
ends at the end, so the list stayed detached and later output grew below
the viewport. The release now clamps the tapped distance to the room
left. A row hold also clears an end hold still in progress, so the plan
card or another toggle anchors its own row.
* fix(app): keep following when a reply lands during a final prompt collapse
The end hold clamped its distance from the end only against the final
content size. A reply that grew the content after a collapse reached the
end brought the old distance back, so the hold released detached. The
hold now clamps on every content size change.
* fix(app): let a reply grow below a held final prompt
An end hold pinned the content end for the whole hold, so a reply that
landed during it scrolled the toggled prompt up. The pin now holds only
while the toggled row is the last row, and drops in the commit that
mounts the reply.
* fix(app): move a waiting final prompt hold to a reply that lands first
The end hold's toggle waits a frame for the end pin, and it kept the
anchor from the tap. A reply landing in that frame dropped the pin but
left the hold with no row, so the toggle moved the button. The disclosure
hold now owns the waiting toggle and resolves the anchor when the frame
runs.
#8740 moved the workspace header into every split pane so each pane could
pick full or compact chrome by its own width. Every pane then repeated the
title, project row, workspace menu and Explorer toggle, and narrow panes
swapped their tab rail for the tab counter.
The tab counter belongs to the single-pane layout. With two or more visible
panes, the split container now renders the header once above them (full or
compact by column width, without a counter) and every pane keeps its tab
rail. A single visible pane, including focus mode, keeps the merged compact
header with the counter when narrow.
* fix(client): recover a dead connection within seconds of returning to the foreground
A socket that died while the OS had JS suspended stays OPEN and keeps accepting
sends, so nothing but inbound frames can prove it alive. Since #8651 the
foreground check rode the regular heartbeat: a 15 s window, dedupe onto a probe
that straddled the suspension, misses from before background carried over, and
a 1.5 s backoff after teardown. Measured at the transport: 40 s to tear down and
41.5 s to reconnect, with 15 s to 30 s variations depending on what was in
flight when the app was suspended.
Foreground now starts a fresh cycle inside the same heartbeat owner: two 5 s
windows back to back, then teardown and an immediate reconnect. An outstanding
heartbeat probe is dropped without a miss because it measured the suspension,
not the socket; a latency measurement keeps its own deadline and is shared.
Repeated foreground events join the running cycle instead of restarting it.
The rule that any validated inbound frame voids a window is unchanged, so a
slow multi-frame transfer still survives, and after any progress the heartbeat
returns to its 10 s / 15 s cadence. Ordinary heartbeat teardowns keep their
backoff; only the foreground cycle reconnects at once, and a failed immediate
attempt backs off as before.
A single frame that takes longer than the two windows to arrive after
foreground is still torn down. That needs smaller daemon-to-client frames and
is documented as a known limit.
* fix(client): count only fresh evidence in the foreground liveness cycle
Two gaps in the foreground cycle let stale evidence decide a fresh check.
A latency measurement outstanding at foreground is shared by the check. If a
frame had arrived during that measurement before the app was suspended, the
check read that old activity as "alive" and went back to the normal cadence
without running its windows. The check now clears the shared probe's activity
flag when it begins, so only frames that arrive from then on count.
If the app was suspended again inside the second window, the next resume
either found the queued expiry already fired (teardown before AppState) or
deduped onto the old probe and let the expiry fire right after (teardown with
no fresh window). A foreground window whose timer fires well past its deadline
on the wall clock spanned a suspension; it is now discarded together with any
activity it saw and the misses before it, and the cycle starts over. The wall
clock is used because the monotonic clock may not advance while the OS has the
process suspended.
* fix(client): keep latency measurements alive when foreground replaces a heartbeat
A latency measurement that found a heartbeat ping outstanding joins that
probe. The foreground check used to fail the probe outright to start its
fresh window, which rejected the measurement before its own deadline on a
healthy socket; the diagnostics collector treats that as a failed
collection. The superseded probe is now detached without settling and
takes the outcome of the fresh window, so the heartbeat tick and any
measurement waiting on it see the next pong or the window's timeout
instead of a spurious error.
* fix(client): let latency measurements wait for the next pong instead of a probe
A measurement that joined an outstanding heartbeat probe took that
probe's fate. The foreground check replaces such a probe, so the
measurement failed early: first with the supersede error, then, after
the hand-off to the fresh window, with that window's five-second timeout
even when the caller had asked for a longer deadline and the pong was
about to arrive.
Measurements now register as pong waiters. A waiter settles on the next
pong, on a connection failure (close, transport dispose, ping send
failure) or on its own deadline, and no longer cares which probe put the
ping on the wire. The foreground check goes back to dropping the
straddling heartbeat probe outright, since the only thing waiting on a
probe's promise is the heartbeat tick.
* fix(client): report the joined ping's round trip from a latency measurement
A measurement that joins an in-flight heartbeat ping used to start its
clock at the join, so it reported only the remaining wait for the pong.
The clock now starts at the send of the ping the measurement waits on,
which matches what the heartbeat records for the same pong.
* fix(server): publish attention clears for unloaded agents
Stored attention clears bypassed agent state events and subscription ownership, leaving connected directory observers stale. Use the existing manager mutation and publish path for unloaded agents, and cover independent filtered subscribers through real daemon requests.
* test(server): run attention regressions in subscription CI
Keep full clear-response coverage alongside independent subscriber checks. Allow asynchronous native deletion classification beyond the default one-second polling deadline.
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Archiving removes an agent from the active directory without deleting the conversation. Preserve its local composer draft there too, and clear it only on explicit entity removal.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
At widths beyond the chat's max width, the latest turn's footer sat at the
chat's left padding instead of under the message column, while running and
after it ended. Earlier footers were aligned.
#8358 wrapped the web viewport's live auxiliary (the latest turn footer and
pending permissions) in a measuring div with no style. That div is a block
box, so the footer row's alignSelf: center no longer applied. Give it the
same flex column style as the stream rows.
Fixes#9081
The completed turn footer wrapped "Worked for 26s" onto two lines at full
chat width in builds before #9007, which sized the duration to whichever
label had more characters. A label with fewer characters can paint wider,
so the duration wrapped inside the narrower timestamp's width. #9007 gives
both labels the full label column; this locks that behaviour with a browser
test using real layout.
The browser test project could not load anything that reaches the toast
context: react-native-web has no ToastAndroid, and native ESM rejects the
named import. A react-native test shim re-exports react-native-web and adds
it. The lucide test stub gains the icons the message module imports.
* fix(app): keep one hover card open while scanning sidebar workspaces
Sweeping the pointer down the workspace list showed cards about three
frames late, stacked two or three at once, and left some open after the
pointer moved on.
The card was positioned through an async chain (measureInWindow, then
onLayout, then a position effect, then the portal's passive mount), so it
painted at -9999 first. While unplaced, its safe-zone bridge covered
everything above the trigger, so upward pointer moves kept cancelling its
close timer. The previous card also stayed for the close grace period
plus its fade while the next one opened.
Now each overlay layer holds one open card, and opening another trigger
closes it immediately. The card reads its trigger and overlay rects in a
layout effect inside the portal, so the first painted frame is already
placed. The grace timer still covers jitter and the gap onto the card.
* test(app): name the hover scan workspaces so the resting row's card is identified
The runtime already identifies connections through another host as routed, but both diagnostic fields fell through to unknown. Preserve that distinction in the report.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* feat(protocol): define streamed diff revisions and resolve tag bases
* feat(server): stream shared diff revisions with paced file bodies
* feat(client): observe streamed diffs with atomic revision updates
* feat(app): show the Changes tree while diff bodies stream
* fix(server): retain the legacy structured diff payload budget
* fix(diff): preserve revision ordering and readable files
* fix(diff): select renamed patches and read dangling symlinks
* fix(diff): preserve nested checkouts and ambiguous patch paths
Only regular files and symlinks are readable untracked entries. Prefer patch metadata to ambiguous unquoted headers so paths containing a prefix delimiter keep their hunks.
* fix(diff): preserve BOM text and complete file replacements
Decode each string without consuming its leading BOM. File-type replacements emit multiple patch sections for one listed path, so retain all matching hunks. Assert the optional rename metadata in the existing commit-file journey.
* fix(diff): highlight symlinks from their stored target text
Working tree source context must match the symlink blob emitted by Git rather than the contents reached by following the link. Otherwise highlight runs do not cover the patch line and encoding fails the revision.
* fix(diff): tolerate disappearing highlight sources
Source context is optional after Git has produced the patch. Missing worktree link metadata must preserve the existing hunk-based highlighting fallback instead of failing the revision.
* fix(diff): batch cold reads and refresh logical comparison inputs
Acquire bounded pages of tracked patches lazily after Begin while retaining per-file highlighting and awaited frame delivery. Keep legacy preferred-base selection live across revisions and include submodule Git state in worktree cache identity.
* fix(diff): distinguish gitlink directories from replacements
A raw old gitlink mode also occurs when a submodule becomes a regular file. Query nested Git state only for a current directory, preserving normal file identity for replacements.
* fix(diff): invalidate timestamp-preserving file replacements
Size and modification time do not identify live file contents after a preserving copy. Include filesystem change time and file identity for tracked and untracked reuse, and verify live revisions agree with fresh reads.
* fix(diff): align highlight tokens with normalized patch lines
Git can normalize CRLF worktrees while highlighting reads their original line endings. Match tokens to the emitted patch text and trim only a normalization CR before encoding runs; mismatched source tokens are not attached.
* fix(diff): refresh equivalent worktree paths across platforms
Explicit refreshes must match filesystem-equivalent roots, including Windows path aliases. Use the existing realpath-aware matcher, budget real Git completion waits consistently, and assert optional status metadata in the filesystem commit-file journey.
* fix(diff): validate Git inputs and resolve refresh roots
Reuse now accounts for effective attributes, configuration, and canonical worktree blobs as well as filesystem state. Explicit refresh resolves repository identity like subscription creation, and callers await that resolution before acknowledging the command.
* fix(diff): keep non-Git workspace refresh successful
Diff invalidation has no target repository for an ordinary folder. Treat that expected absence as a no-op while preserving other root-resolution errors.
* fix(diff): observe scoped Git interpretation settings
Classify each metadata root with its worktree and common scopes, retaining both when the directories coincide. Shared attributes affect all checkouts, while config.worktree belongs to the owning checkout. Real subscriptions verify automatic binary and hunk-context refreshes in main and linked worktrees.
* fix(diff): observe interpretation inputs for pinned comparisons
Immutable commit objects still depend on mutable Git configuration and attributes. Declare those metadata inputs independently of ref, index, and worktree mutability, while keeping unrelated endpoint events out of pinned comparisons.
* fix(diff): observe interpretation independently of comparison endpoints
Commit comparisons still read worktree attributes. Watch those inputs independently, checking effective Git interpretation before invalidating fixed contents. Retain paths hidden by whitespace filtering so attribute changes can expose them. Forward binary frames through the branch-loading test interceptor.
* fix(client): close abandoned browser connections immediately
When the client gave up on a connection (liveness timeout, connect
timeout, reconnect, subscription failure) it closed the WebSocket with
1001. Browsers, Electron and Node's global WebSocket accept only 1000 or
3000-4999 and throw InvalidAccessError otherwise; the throw was
swallowed and the reference dropped, so the physical socket stayed open
and the daemon kept streaming into it until its 45 s lease expired.
The WebSocket adapter is the only code that calls WebSocket.close, so it
now sends 1000 for codes the API rejects and trims reasons to the
123-byte limit. Callers keep their reasons, and Node's `ws` sends the
same frame as a browser, so Node tests now catch this class of bug.
* test(client): drop timing-dependent hello count from abandoned-close test
Handshake retries after close are owned and tested by the relay's handshake-lifetime test; this test only checks that the abandoned socket closes.
A missing runtime timestamp does not establish connection history. Report it as unknown while preserving supplied timestamps.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Read the submitted form value while hydration state is still catching up with text entered before its input listener. Cover the missed input event without relying on startup timing.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
The server-rendered trigger accepted clicks before its handler existed. Tie its enabled state to hydration and cover delayed scripts so an early click cannot silently disappear.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* fix(app): preserve host error details in desktop logs
Bind host context through the existing client logger and emit structured text so desktop console capture retains the error record, including for adopted probe clients.
* fix(app): retain discovered host identity and nested error causes
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Consume host connection statuses in History instead of a version-only dependency that the React Compiler removes. Keep saved sessions reachable when the host connects after the screen mounts.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Cancel handshake work at the channel close transition instead of waiting for a socket acknowledgment from an unreachable peer.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
The upgrade journey still expected protocol 5 after the worker moved to 6, stopping every main CLI run before adoption coverage. Keep obsolete versions 0 through 4 and exercise newer-worker rejection with version 7.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* fix(desktop): fill controlled form fields before submission
Write text controls through native DOM setters so page-level value tracking sees the input event as a change. Cover state updates, submission, replacement and clearing through the real browser tool journey.
* refactor(desktop): select native field setters with explicit branches
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Reopening a chat lists its commands, which resumes the agent and starts
a Claude process with no prompt. Claude only sends init with a turn, so
the runtime's background-work state stayed unknown and idle cleanup
vetoed it on every sweep: the process stayed resident until the daemon
restarted.
A process that has not been sent a prompt and has not reported init has
run no turn and owns no background work, so it can stop. Older CLIs
that ran a turn without a roster are still kept.
Fixes#9017
Co-authored-by: Brandon Elliott <923787+Brandon168@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Pointer submission blurs the editor before sending and clears retained focus intent. Restore focus through the existing default send action for mouse-driven layouts, while keeping touch keyboard dismissal intact.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* fix(app): fill short collapsed chat histories
Collapsed whole turns can leave the initial viewport short even when older history exists. Continue paging through measured pages before revealing a fresh chat, while preserving visible history and the reading anchor.
* fix(app): settle padded and stalled chat history
Measure the native history edge using the same content area as viewport filling, and bound child history requests so retry can start after a stalled fetch.
* fix(app): adapt viewport filling to the current timeline
* fix(app): resume history filling after switching tabs
* fix(app): make a folded turn's summary line its own chat row
The summary line of a collapsed turn was drawn inside its prompt's row, so the
row gap landed below the summary and the prompt touched it. Turn collapse now
emits a `turn_summary` row between the host and the reply, and one spacing
rule covers every prompt: 16 above whatever answers it (reply, summary row or
tool row). The summary row spaces like the work rows it stands for, 4 above the
reply and flush under a tool host.
This removes the per-row summary lookup, the host revision tracking and the
fragments in the view. A fold reuses a turn's previous summary row object while
its work and open state are unchanged, so mounted rows do not re-render.
* fix(app): keep the turn summary row at its host's timeline position
The chat outline resolves the active prompt from the timeline position of
the row under the reading line. The summary row had none, so the outline lost
its mark while the row sat there. The row now carries its host's position and
is replaced when that position lands.
* fix(app): keep the reader on a folding turn's host, not its summary row
When an older page completes the oldest loaded turn and it folds, the reading
position moves from the hidden rows to the turn's host row, as on main, so
the prompt and its summary row both land in view. The pagination helper now
finds the summary line in the row under the prompt.
* fix(app): keep turn footer timing within the chat width
Let visible timing text determine its height within the available chat width instead of sizing an absolute overlay by character count. Keep actions anchored when duration and timestamp labels swap.
* fix(app): keep footer timestamp hover target stable
* fix(app): reserve both wrapped timing label heights
* Serialize Codex app-server startup by state directory
* Keep Codex session constructor unchanged
* Type Codex startup failures and inject process launch port
* fix(codex): retain startup diagnostics emitted after exit
Classify failed initialization after stdio closes so the bounded SQLite retry sees the complete stderr. Keep running-session exit handling immediate. The regression delivers the SQLite diagnostic between exit and close.
* test(codex): inject history test process through the public dependency
The history fixture still replaced the old private spawn method, so provider launch preparation tried to find a real CLI in CI. Inject the same process interface as other provider tests and use an invalid command as a guard against accidental real launches.
---------
Co-authored-by: Michael Wu <michaelmwu@gmail.com>
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Ivan Malison <1246619+colonelpanic8@users.noreply.github.com>
* feat(plugin): expose stable sidebar entity data across hosts
Collections return interned references and preserve their arrays when membership and order stay unchanged. Entity selectors share one hook with identity defaults for full snapshots and shallow defaults for selections. The app caches host projections and shares underlying store and host subscriptions. Browser render counts exercise the real session store and plugin provider without memoized rows. Terminal hooks consume the collection source port; daemon-wide terminal discovery is the separate terminal phase.
* feat(plugin): let users select a sidebar view
An exclusive view replaces the sidebar content while the host retains navigation, callouts, and footer. The device remembers the selection, unavailable views fall back to the default, and throwing views provide recovery. Device-scoped plugin settings persist locally with schema validation and revision conflicts. Web E2E and Android tablet QA cover selection, preserved chrome, disabled-view fallback, and error recovery.
* feat(plugin): add sidebar view examples and ref actions
Keep sidebar layout and gestures inside the plugin while routing entity handles to their owning hosts. Examples cover empty projects, pinned workspaces, task sections, nested agents and device-local label folders.
* feat(plugin): observe terminals across connected hosts
Share host-wide terminal observations across plugin surfaces and gate them on terminalCollection. Keep directory observations compatible, and carry creation time in worker v6 state so v5 terminals can drain with an unknown timestamp.
* fix(app): preserve linked resources when pasting prompts
Rich clipboard links carried their destination only in HTML, while the composer let textarea default paste insert the plain label. Convert linked fragments to Markdown at ingestion and replace the selected draft range through the existing editor command.
* fix(app): keep pasted links undoable while drafting offline
Use the browser edit command for converted paste so selection replacement participates in Undo and normal input publication. Keep complete Markdown clipboard text when it already carries the link destinations, and limit connection gating to image uploads.
* fix(app): preserve copied Markdown with normalized links
Copy renders normalized URLs in HTML while retaining original Markdown in plain text. Compare parsed destinations using the same Markdown parser so pasting complete responses keeps tables and formatting intact.
* fix(app): retain external linked tables when pasting prompts
Recovering missing link destinations must retain table rows and columns. Use the existing table converter and keep cell delimiters and line breaks inside their cells.
* test(app): name rich clipboard paste arguments
* fix(app): preserve pasted links around rowless tables
The table converter requires a first row. Convert rowless table content separately so empty markup cannot abort recovery of surrounding destinations.
* test(app): wait for workspace restore before undo teardown
The directory appears before git finishes recreating the worktree. Wait for the host to publish its active workspace before cleanup removes the source repository.
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* fix(desktop): run browser_evaluate on pages whose CSP omits 'unsafe-eval'
* test(desktop): browser_evaluate on pages that allow and forbid eval
* test(desktop): e2e opens a page under a CSP without 'unsafe-eval' and evaluates on it
* fix(desktop): preserve page globals in browser evaluation
Define the supplied function outside every wrapper binding so strict-policy evaluation keeps the original page scope.
Co-authored-by: Abdulhakeem Mustapha <abdulhakeemmustapha@gmail.com>
---------
Co-authored-by: Abdulhakeem Mustapha <abdulhakeemmustapha@gmail.com>
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* fix(server): keep the model selected by the provider
A live model setter can retain or resolve a different model than the requested ID. Read its effective selection before persisting manager state so an ignored selection cannot overwrite the actual model. Preserve explicit preference resets.
* fix(server): preserve completed model selections
Return the effective selection from the provider mutation so a separate status request cannot reject an already completed switch or leave configured and runtime models inconsistent.
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Snapshot refresh completion does not mean both workspaces have joined the repository fetch target. Wait for observation setup before releasing the first fetch, and retain delayed checkout resolution to exercise that ordering.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* fix(desktop): browser_logs reports the HTTP status of each request
* test(desktop): browser_logs reports the status the browser recorded
* test(desktop): e2e checks browser_logs reports a 404 and a 200
* docs(browser): browser_logs also reports the HTTP status
* fix(server): preserve the selected model when importing sessions
Import preparation resolved a new-session catalog default and sent it as an explicit resume override. Keep import selections with the resumed provider and persist its reported model for subsequent reloads.
* fix(codex): allow the first reply after importing a session
Resolve collaboration settings again when default model and effort are applied so imported sessions send a complete first-turn request.
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* M1: terminal input-sink bridge, upload button in tab-bar slot and key bar, picked-file routing, space-separated pasted paths
* M2: terminal dictation: mic in tab-bar slot and key bar, recording pill and overlay, per-terminal recording lifetime incl. pending mic acquisition, Cmd+D on macOS by keyboard focus, i18n
* M2: review fixes: retry attempt guard, cancellable native pending start, measured terminal slot width, compact-web Cmd+D for terminal dictation
* fix(terminal): separate uploaded file paths when pasting
Keep the remote file-drop regression and its test coverage from the contributor change, without introducing new terminal controls. Each uploaded path needs a trailing separator because adjacent quoted paths are one shell word.
Co-authored-by: Sebastian Arena <sebastian@walkofcode.io>
---------
Co-authored-by: Sebastian Arena <sebastian@walkofcode.io>
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Mohamed Boudra