Failing reproduction for the third review finding on #10179. The
frame-budget fallback returns item pages whose start cursor can fall
inside the earlier turn. A whole-turn `before` page from that cursor
extends to the end of the turn that holds it, so it ends at the cursor
instead of the row before it, and the app drops the page.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
d016bf0b4 stopped the frame-budget fallback when a reselect returned as
many rows as the previous one. A turn whose spanning tool call completes
a few rows before the end returns the same rows for limits that land
inside the span, but a smaller limit lands after it and fits. The
fallback now halves the limit down to 1 and picks the error from the
row count of that last page.
Co-authored-by: AnalyticETH <hello@analytic.lol>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A tail page starts at the first row of the source span it lands in. When
one tool call starts before an oversized turn and completes after it, its
span covers every row, so a smaller limit returns the same page. The
fallback halved the limit down to 1 and then reported that a single item
was too large. It now stops at the first reselect that does not shrink and
says that one item spans more than the limit.
Document in the SDK reference, the plugin hook docs, and the zh-CN copies
that a whole-turn page over the message limit arrives as an item page that
can start partway through a turn.
Co-authored-by: AnalyticETH <hello@analytic.lol>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Opening a chat whose newest turn is larger than the 100 MiB message
limit never showed its history. The daemon built one reply for the
complete turn, the outbound transport refused it with "Logical message
too large", and the app waited until its request timed out.
This test fills one turn with about 105 MiB of replies and reads it as
the app does: a whole-turn tail page of 40 items, then older pages
until the start. On main the first request gets no reply within 20 s.
With the frame-budget fallback, every reply arrives and the pages reach
the start of the turn. The test runs in test:integration so CI covers it.
Refs #8458
Co-authored-by: AnalyticETH <hello@analytic.lol>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Mermaid 11's lexer reads an edge id (`e1@-->`) from any run of
non-space characters ending in "@". In `F[@mention] --> G` that run
starts at the node id and crosses the label's opening bracket, so the
parse fails and the card falls back to the source. Backport the lexer
half of mermaid-js/mermaid#8459, which stops the edge id at the node
shape characters, onto the pinned mermaid 11.16.0 and rebuild the
runtime. Moving to mermaid 12 instead changes the default layout and
look of every diagram and raises the browser floor, so it stays a
separate decision.
Closes#10157
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: monarchmaisuriya <37824914+monarchmaisuriya@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
When a plugin provider closed a session with session.closed, the next
prompt went stale and Paseo tried to reopen the agent. The reopen first
closes the old session, and that close sent session.close for a session
the plugin runtime had already forgotten. The send failed with "Unknown
provider session", the reopen stopped, and the prompt was rejected with
"Provider session <id> is stale after its plugin reloaded". Every later
prompt failed the same way.
A session that the provider closed now skips the native close and only
drops its routing, so the reopen continues and the prompt runs.
Closes#10151
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
"forks a streaming assistant turn without interrupting it" pressed the
in-flight turn's fork trigger once and waited for the menu. While the
chat is shorter than the viewport, each streamed line moves the trigger
down. A line that arrives between the press and the release moves the
trigger out from under the pointer, the press opens nothing, and the
wait for "Fork in a new tab" timed out.
openForkMenu now presses the trigger until its menu shows.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
When the app opens straight into a chat on a host it hasn't connected to yet,
such as a page reload or a cold start on an agent route, it renders the cached
messages before the host's client exists. useMarkdownSource enabled the image
read anyway, with no client to read through, so it failed with "Image
unavailable". The query's retries cover a host that connects within about
seven seconds. After that the read stays failed and the image collapses to its
alt text. The query key doesn't include the client, so the client's arrival
never starts another read, and the image stays as alt text after the host
connects.
Markdown that belongs to a host, which has a serverId, now waits for that
host's client before it reads. Until the host connects the image keeps its
pending placeholder, and once the client arrives the query runs and reads the
file. Markdown with no host, such as the changelog, has no client to wait for
and still falls back to its alt text.
* fix(omp): map tool approvals that offer session or grant choices
OMP 18.x adds "Approve <tool> for this session" and directory/file grant
choices to its rpc-ui approval select. The mapper only recognised the exact
["Approve", "Deny"] options, so these prompts fell through to a generic
"OMP select" question; allowing it from the CLI or a notification answered
with no value and OMP treated the call as denied.
Recognise approval selects whose one-call choice is "Approve" or
"Approve Once", whose last choice is "Deny", and whose other choices are
approval labels. Surface the extra choices as allow actions and answer with
the selected label.
Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
* fix(omp): allow one call when a grant is labeled Approve
OMP names the one-call choice "Approve Once" whenever it also offers
grants. A grant for a target that is not a file is then labeled
"Approve", so the select arrives as
["Approve", "Always Approve", "Approve Once", "Deny"]. Taking the
first "Approve" as the one-call choice made a plain allow pick the
grant, which remembers more than the call. Prefer "Approve Once" when
it is offered.
Co-authored-by: Dhruv Singal <dhruvsingalabc@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* refactor(omp): read the picked approval choice from its action
Each extra approval choice's action label is the select value OMP
offered, so the response reads it from request.actions instead of a
second copy in metadata.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Dhruv Singal <dhruvsingalabc@gmail.com>
Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Use current Hub permissions when reconnecting a daemon
* Back off when the Hub repeatedly rejects session permissions
* Show the permission recovery commands in Hub status
Supported providers is now an index: one list of natively supported providers and one list of ACP agents from the in-app catalog. Each natively supported provider has a setup page under /docs/supported-providers/, and the docs sidebar shows those pages as one collapsible item. /docs/claude-code, /docs/codex, and /docs/muse-code redirect to the new location.
Also adds landing pages for fx and Hoplite, lists fx in the 0.16.0 changelog highlights, lists every docs page in llms.txt, and gives the plugin sidebar views page its frontmatter.
The sidebar toast floats over the list's end, so at full scroll it covered the
last row. The list now ends with a spacer the height of the toast while one
shows. The spacer sits below every row, so growing it moves nothing. It shrinks
over the toast's fade, so a list scrolled into that space settles back with the
toast instead of snapping.
Unlike the permanent room removed in #9667, the space exists only while a toast
is up, so the list keeps its normal end the rest of the time.
The workspace-labels journey hovered a row once and then waited for its
kebab. The kebab mounts only while the row holds hover, and hover only
arrives on pointerenter, so a row that lost hover under a parked pointer
never showed it again and the click waited out the 3-minute test timeout.
openWorkspaceSidebarKebab now moves the pointer off the row and hovers it
again until the kebab is visible, and the labels journey uses it instead
of its own copy.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
* fix(app): preserve literal text in selection copy
* fix(app): keep literal angle brackets in copied selection HTML
Unwrapping a recognized URL left displayed text such as
`<http://localhost:3000>` in the Markdown that feeds text/html. Turndown
does not escape `<`, so the clipboard renderer turned it into an
autolink and dropped the brackets. Escape `<` in the Markdown flavor.
Describe the renderer/copy link contract in docs/markdown-renderer.md
instead of restating the code.
Co-authored-by: Dan Berindei <dan.berindei@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(app): keep recognized web links and table pipes in rich and plain copy
Unwrapping recognized links in the shared copy DOM removed them from
both clipboard flavors, so a copied `#12` lost its forge URL in rich
paste. The renderer now marks recognized links with their own
attribute: text/plain carries them as the text the reply showed, and
text/html keeps them as links when they point to a web page. Workspace
paths stay plain text in both.
text/plain still emits tables as GFM syntax, so keep the cell-pipe
escape in both flavors.
Co-authored-by: Dan Berindei <dan.berindei@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(app): copy recognized Windows drive paths as text in rich paste
The rich-paste link check accepted any URI scheme, so `C:\repo\x.ts`
kept a link with an unopenable `C:` href. Keep recognized links in
text/html only for http and https URLs.
Co-authored-by: Dan Berindei <dan.berindei@gmail.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(app): keep split line-start Markdown literal in copied selection HTML
Selecting from the `#` in `**C#**` to the end of the paragraph left `#`
and ` for ...` as separate text nodes after unwrapping, so Turndown did
not escape the heading marker and rich paste showed a heading. Merge the
text nodes before serializing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Dan Berindei <dan.berindei@gmail.com>
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Keep a finished reply visible when Claude starts a turn after a background command
When a background command finishes while Claude is idle, Claude Code sends the
task notification first and opens its own turn after it. The notification row
was stored with no turn id. A row with no turn id joins the turn before it to
the turn after it, so under Collapse turns the app showed both turns as one and
folded the finished reply behind the summary row.
The Claude provider now keeps a task notification that arrives between turns
and emits it inside the turn Claude starts for it, so the row carries that
turn's id. A notification no turn follows is emitted as before: when Claude
reports idle (a command the user stopped), before the user's next message, and
on close.
Rows already stored without a turn id are not changed.
Refs #10030.
* Show a stopped background command's notification at once
A stopped background command starts no Claude turn. While a background
subagent is alive Claude also sends no idle, so the held row stayed hidden
until the subagent finished and then took the subagent's wake-up turn id.
Only task notifications Claude reacts to (completed or failed) wait for a
turn now. A stopped one is emitted right away, as before.
Refs #10030.
* Keep a row that belongs to no turn out of the turns around it
A turn starts with the user's message or the provider's own turn start
(Claude's init). A background command's notification that arrives
between turns starts nothing, so the daemon writes it with no turn ID.
The shared turn rule read a missing ID as legacy history and let the
row continue both neighbours. The reply before it and the turn after
it then read as one turn, and Collapse turns folded the first reply.
Now a row with no turn ID never continues a row that has one, so it
cannot join two turns. History with no IDs at all keeps the trigger
rule. The app's fold and the daemon's whole-turn paging share the rule.
This replaces the earlier Claude provider hold, which had to guess from
the notification's status whether Claude would start a turn for it.
Refs #10030
* Give a background task's notification the turn it starts
A background task's notification that arrives while no turn runs starts
a turn. Claude's next run belongs to it unless Claude announces another
start first: a later notification starts the next turn, and a queued
command (the user's message, a scheduled wake-up) starts its own. When
Claude does not respond, the notification is a turn in itself.
So a reply Claude writes for a finished background command shares the
notification's turn, and the reply before it stays in its own turn and
stays visible under Collapse turns. A stopped command's row shows at
once as its own turn, and the agent stays idle.
Claude sends no frame for a Monitor event, so a run it starts for one
belongs to the last notification before it.
This replaces the shared turn-rule change, which drew the notification
as a turn between the two.
Refs #10030
* Keep a subagent's own job notification out of the main session's turns
A background job a subagent started reports to that subagent, and
Claude starts no main-session run for it. Its notification gave the
next main-session run a new turn, so when it arrived between a main
command's notification and Claude's reply to it, the reply was drawn
in a different turn from the notification it answered.
Refs #10030
When OMP changed the thinking level itself, for example on `/switch @role`
with a role that carries a level, Paseo dropped the `thinking_level_changed`
event because the OMP runtime schema did not list it. The thinking selector
and `runtimeInfo.thinkingOptionId` kept the old level while OMP ran the new
one.
Parse the event and update the session's level, preferring OMP's
`configured` selector (for example "auto") over the effective level, the
same way the Pi provider handles it (#6741).
Closes#8249
Co-authored-by: factory-droid[bot] <138933559+factory-droid[bot]@users.noreply.github.com>
An in-app update and the startup version check again use the desktopManaged flag in paseo.pid, while Manage built-in daemon is on. A daemon that an earlier Desktop session launched is now replaced on update. A daemon that the user started through the CLI is still never stopped automatically.
Closes#6385.
* fix(android): keep composer taps working after pinching an image
React Native counted two touches at the start of a pinch but received a native takeover cancellation for just one. Typing could then claim the responder and consume the first toolbar tap. Cancel the active contacts when the gesture handler takes over.
* docs(android): describe the chat image fixture for the pinch regression
Give worktree teardown commands a shared one-minute deadline. Projects can override it with `worktree.teardownTimeoutMs` in `paseo.json`. Drain output so successful teardowns can exceed the old 1 MiB buffer. Bound the output retained for diagnostics.
At the deadline, try to terminate the command, close its output streams, and keep the worktree on disk. An exit status of zero in response to the stop signal still counts as a timeout.
Closes#9857.
Co-authored-by: Okan YILDIRIM <63511519+ragokan@users.noreply.github.com>
Recover missing timeline rows when a chat was empty before disconnect and live reply rows arrive before history catch-up. Keep the replica cache at schema version 1.
Closes#9650.
Supersedes #9626.
Co-authored-by: Okan YILDIRIM <63511519+ragokan@users.noreply.github.com>
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Viewport and full-page browser screenshots wait for a frame with non-transparent pixels. When the page does not paint within the existing capture budget, the tool returns screenshot_no_frame.
Fixes#9810.
Co-authored-by: monarchmaisuriya <monarchmaisuriya7600@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
List ignored directories as single entries and only search inside them
when a .worktreeinclude pattern matches the directory or names a path
inside it. A bare .env no longer selects node_modules/pkg/.env.
Verified with real Git and isolated daemon workspace creation on Linux.
With 300,000 ignored files, Git trace2 recorded over 6,000 directory
visits before the fix and fewer than five after. The workspace copied
the root .env. The lifecycle test also verified rollback when copying
fails.
Closes#10004
Co-authored-by: mgranderath <22001111+mgranderath@users.noreply.github.com>
* Tell plugin agents to develop in a directory plugin, never a Git or npm installation
paseo plugin update replaces a Git or npm installation and remove deletes it,
so edits and commits made in Paseo's managed copy are lost. The paseo-plugin
skill now says to clone the source outside the daemon home and install that
directory, and to move any work out of a managed copy before update or remove.
Refs #7904
* Rewrite the plugin development guidance as plain steps
* Keep settings, runtime ID, and build output when moving a plugin to a local folder
* Keep background workspaces out of the dock badge and favicon
A workspace created with `--background` is hidden from the sidebar, but an
agent finishing in it still counted toward the macOS dock badge and turned
the web favicon to attention, pointing at something the user cannot see.
The dock badge now skips background workspaces, and the favicon skips
agents whose workspace is background, the same way both already skipped
background agents. The favicon selector also rescans when the workspace
map changes, so an agent synced before its workspace is dropped once the
workspace arrives.
Closes#9975
* Count background workspaces in the badge while Show background is on
The sidebar shows background workspaces when Show background is on, so
the dock badge and favicon now count them in that case too. They follow
the same setting the sidebar reads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Let plugin screens use the plugin state hooks
A plugin screen opened from the Command Center or the sidebar showed
"Plugin failed: Plugin state hooks must run inside a plugin host" as soon
as it called useWorkspaces, useAgents or another state hook. Every plugin
host paired the installation provider with its own state provider, and
the screen host and the plugin settings host did not.
The installation provider now supplies the state source for the
installation's host, so every surface rendered under it can read state,
and the hosts no longer wrap a second provider of their own.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Keep a sidebar view's cross-host state in its hooks
A sidebar view renders once across every connected host and builds an
unscoped state source for that. After the installation provider took
over the state source, the environment provider dropped the view's own
source, so its hooks read state scoped to one host. The installation
provider now takes the surface's source when one is passed and falls
back to the installation's host.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(terminal): delete a stopped worker's copy on Windows
Windows keeps a just-exited worker's files open for a moment, so the
single rm in pruneTerminalWorkerRuntimes failed and left the copy until
the next daemon start. Retry the delete for up to 5.5s.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(terminal): bound the Windows retry of a worker copy delete
rm's maxRetries repeats at every directory level, so a file that stays
locked deep in node-pty would hold daemon start and stop far past the
intended window. Retry one rm per attempt for up to 3s instead.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(app): publish restored hosts together with their directory sync
A component that rendered while the host registry restored the project
icon cache saw the host but found no directory sync for it, so its
directory demand was dropped and the sidebar stayed on No projects yet
for the session. Hosts are now published in the same step that creates
their runtimes.
Fixes#9996
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* refactor(app): publish restored hosts through setHostsAndSync
Every host-list change now goes through the one method that assigns the
hosts, creates their runtimes, and notifies listeners in a single step.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(workspaces): preserve commits when archiving worktrees
A removed worktree can be the only reference to its HEAD. Keep that commit before deletion and record the checkout so detached workspaces restore at the same commit.
Co-authored-by: Okan YILDIRIM <63511519+ragokan@users.noreply.github.com>
* fix(workspaces): restore saved checkouts through one git path
Keep existing archive and deletion behavior when Git cannot read a worktree HEAD, and record every readable checkout for a common restore path.
* fix(workspaces): preserve git HEAD identity through restore
Record Git’s full HEAD ref or commit so restore can pass branch and detached checkouts to Git without guessing from a short name.
* fix(workspaces): scope saved HEAD to each archive
An archive with no readable checkout must not retain the HEAD saved by an earlier archive.
* test(workspaces): verify detached checkout after restore
Update the browser journey for detached restore and allow the daemon restart test enough time on Windows runners.
---------
Co-authored-by: Okan YILDIRIM <63511519+ragokan@users.noreply.github.com>
* Reload the desktop window when its renderer exits
When the renderer process of the main window exits without a clean
shutdown (killed by an out-of-memory killer, crashed, or out of memory),
Electron keeps the window frame but shows nothing until the user presses
Ctrl+R. The main process had no handler for render-process-gone, so it
neither recovered nor logged the exit.
The window now reloads itself and logs the exit reason to main.log. If
the renderer exits three times within a minute, it stops reloading and
logs that, so a page that crashes on load does not restart in a loop.
Fixes#9999
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Leave a renderer exit before the first page commits to Chromium
Before the window's first navigation commits, Chromium finishes it in a
new renderer by itself. Reloading at that point aborted the initial
loadURL with ERR_ABORTED, which fails desktop startup. Skip the reload
while the window has no committed URL.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Let a steer wait out a Claude compaction instead of cancelling it
While Claude compacted, the Claude session answered every steer with
"unavailable". The agent manager reads that as "replace the turn", so
it interrupted the turn and Claude Code printed "Compaction canceled.".
Claude Code already holds a `next`-priority message until compaction
finishes, so the session now accepts the steer during compaction.
Co-authored-by: NightMachinery <36224762+NightMachinery@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Steer only while the compaction is still running in the real Claude test
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: NightMachinery <36224762+NightMachinery@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The MCP agent tools serialized live agents without the background work
summary that the WebSocket snapshot carries, so a parent agent saw an
idle worker with a running background shell as plain idle. Both tools now
add the same summary the session adds, omitted when the agent has none.
Co-authored-by: Orcun Simsek <1612171+ortschun@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
React Native 0.86 converts between px and dp with the density of the
phone's built-in display, while Fabric lays out at the window's density.
In a DeX window (240 dpi on a 360 dpi phone) every text box came out at
two thirds of the height its text needed, so the text was clipped.
Backport the rule from react/react-native#59087: right after React
Native reads the display (when it creates the root view and on each
configuration change), restore the window's density and keep only the
display's size. Remove once the app runs React Native 0.89 or later
with useWindowDensityForDisplayMetricsAndroid on.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
The plugin-buttons journey reads the details' event subscription id to
check that closing, crashing and disabling the content releases it. The
id arrives with the daemon's subscription response, after the details
render. The first open already waited for it; the reopens read it as
soon as the cached summary showed, so the id could still be empty and
the release poll then waited for an id that never exists.
Every read now waits for a non-empty id.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Changing the thinking level of a running Claude agent marked its query for a
restart. The next turn, mode change or model change then retired the Claude
process, which ended every background shell, workflow and subagent it ran with
"connection lost".
A move between two effort levels now goes to the running query through
applyFlagSettings({ effortLevel }), as Fast mode already does, and marks no
restart. Switching thinking off, to ultracode or back to the default still
restarts, because those change how the query is launched. If the running
process does not take the change, the old restart path runs, so the change is
never lost.
Row: app-built-from-one-set
Paseo: c6210e21-fc13-4dfa-adfc-43a050c7d75a
Co-authored-by: gilad-rubin <6134299+gilad-rubin@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* Create the workspace in the caller's directory when --path is relative
`paseo workspace create --path .` adopted the daemon's working directory
instead of the directory the command ran in. The CLI sent the relative
path unchanged, and the daemon resolved it against its own cwd. The CLI
now resolves a relative --path against the caller's directory before it
sends the request, as it already does when --path is omitted. Absolute
POSIX and Windows paths pass through unchanged so a remote daemon on
another operating system still gets the path as written.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Leave a home-relative --path for the daemon to expand
Resolving every non-absolute --path against the caller's directory turned
`--path=~/repo` (and a quoted "~/repo") into `<cwd>/~/repo`. Before this
change the daemon expanded the leading ~ itself. `~` and `~/…` now pass
through unchanged again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Send a relative --path to an explicit daemon endpoint as written
The CLI already resolves caller paths only for a local daemon and sends
them unchanged to an explicit --host endpoint (resolveRepositoryDirectory,
project create). workspace create, and paseo run, which builds its
workspace source through it, now follow the same rule instead of turning
a relative path into a path on the caller's machine for a remote daemon.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
A force-cancel finalizes turn X in the daemon while the provider keeps
running it, so the next run can be bound to the same X. handleStreamEvent
dropped every terminal for a finalized turn id, including the synthetic
turn_canceled that cancelAgentRunNow dispatches for the tracked run and
the provider's own terminal. The run never settled, cancelAgentRunNow
waited on it forever, and every later cancel, reload and send queued
behind it until the daemon restarted.
A terminal for the tracked run's own turn id is no longer dropped.
Fixes#6355
Supersedes #8405
Co-authored-by: Dominic Anaeto <dominic.anaeto@gmail.com>
Tapping Load more emptied the list until the larger page arrived: the sheet
showed "Loading sessions...", unmounted the virtualized list, and brought it
back scrolled to the top. The sessions queries set keepPreviousData, but
useQueries gives a changed key a new observer with no previous data, so the
placeholder was always empty. The placeholder is now the cached page that
Load more grew from, for the same scope, search, and provider.
The compact import-session journey flaked on this in CI: its scroll helper
resolved a row, the list was replaced before the helper ran, and the walk
up from the detached row found no scroller. The helper now scrolls the list
itself, and the journey checks that Load more keeps the reader down the
list.
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
On web and desktop, Download over a direct connection now fetches the file and saves it from a blob instead of navigating the window to the host URL. A rejected or unreachable download shows an error in the download toast and the app stays on the workspace. The daemon's HTTP CORS check uses the same origin rule as the WebSocket upgrade.
Closes#9895
Co-authored-by: panta82 <4291324+panta82@users.noreply.github.com>
Brings in the shipped 0.16.1 CHANGELOG.md and CHANGELOG.zh-CN.md entries
exactly as tagged, the 0.16.1 version in every package.json and
package-lock.json, and the F-Droid changelog files for version codes
160011-160014.
* test(app): wait for the model control to re-measure in the label sweep
The phone label sweep read the folded model control twice, 100ms apart,
after each resize. A resize reflows the page at once and the browser
clips the model label with its own ellipsis until the control's layout
event arrives. On a slow runner both reads landed in that window, so the
sweep recorded a truncated label beside the thinking label and stopped.
A settled read now also requires the model label to fit its box, and
gives up after 10s with the last state.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test(app): settle the label sweep on the label its box fits
The previous commit rejected one in-between state: the browser clipping
the model label beside the thinking label. A slow runner can also leave
the label truncated to "Ten second…" after the thinking label is gone,
because the label's box widens at once but its text waits for the next
layout event.
A settled read now requires the shown label to match the box: the full
text when the invisible full-text sizer fits, otherwise a truncation
that fits. That covers both states, so the clipped-label special case
goes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(server): treat canceled Codex starts as interruptions
* fix(server): report a replaced pending start as undelivered to its sender
waitForAgentRunStart followed the agent rather than the sender's run, so a
send canceled by a replacement before reaching Codex resolved once the
replacement started. Keep the sender's own pending run and reject when it is
cleared before its turn starts.
The canceled-start tests drain streamAgent, because a fresh Codex agent has no
session handle before its first turn (#8393).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(server): keep a sender's start wait across a stale-session retry
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test(server): expect the start options passed to a foreground run
Session create and scheduled prompts now pass the sender's start options,
so the two call-argument assertions include them.
Co-authored-by: 3ae3ae <3ae3ae@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(server): keep a Codex message canceled before it starts in history
A Stop or replacing send during Codex preparation settled the run as
canceled without recording the sender's prompt, so the message vanished
from history after a reload and on other clients. Record it as a
rejected attempt, as the start-failure branch already does, so it stays
in history without being delivered.
Co-authored-by: 3ae3ae <3ae3ae@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(server): return each send's start from dispatch
A send waited for its turn by looking up the agent's pending run when the
wait began. When a provider plugin's session was stale, the first start
failed and its run was cleared before the sender started waiting, so the
send rejected with "has no pending run" while the retry answered it.
startAgentRun now returns the start of the prompt it dispatched. Every run
created for that prompt, including the stale-session retry, is bound to
it, so the send resolves when its own turn starts and rejects when its own
run ends unstarted. This replaces the retry hand-off between runs in the
agent manager.
Co-authored-by: 3ae3ae <3ae3ae@users.noreply.github.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: 3ae3ae <ogyshjsm@naver.com>
Co-authored-by: Bae Seongmin <83870598+3ae3ae@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: bot <bot@localhost>
Co-authored-by: paseo-bot[bot] <paseo-bot[bot]@users.noreply.github.com>
Co-authored-by: 3ae3ae <3ae3ae@users.noreply.github.com>
Co-authored-by: paseo-bot[bot] <266920839+paseo-bot[bot]@users.noreply.github.com>