* Trace index in the right rail; the trace opens wide beside it
One way in: a Trace section listing the turns by your message; a row goes to the message and opens its turn.
The open trace takes the artifact width with the index as its left column; top-bar button and hover link removed.
Turns are labelled by the message that started them (first_message_id), not by splitting at answers.
* Trace index: the shown turn is marked by its number, not a blue edge
The open trace also keeps the panel's drag-to-resize width when one was set.
* Trace and artifacts can go full width; index rows open up on hover
A stretch icon left of ✕ takes the rail to the whole content area and back, remembered per device.
Index rows stay one line (two in the panel) and run to three lines on hover.
start_worktree creates a branch + worktree by policy and moves the session into it.
Settings card: location, branch prefix, fetch-first, setup command, cleanup on archive.
Rail shows the branch; clean worktrees are removed when the session is archived.
Ships surfaced and enabled beside the everyday Coworker (supersedes the
2026-08-21 lineup for Code only); prompt rebuilt around the discipline of
a worktree per feature, explore-first subagents, proof by tests.
* subagents: durable child transcripts and audit rows
Each child's full message log persists beside the run record (last 50,
purged with the session, served on demand); child tool calls land in the
session audit log stamped with the task id.
* observability: fixes from the review of the recorder feature
Live-step batches retry a failed flush and stay bounded; the Issues checks
run outside the recorder lock and only the ended turn is re-sequenced;
same-millisecond ordering is causal (two flaky tests fixed at the root);
remote automations load their routing before resolving an unknown id.
* gui: plan dock above the composer
The current task's plan rides the composer: collapsed shows the live step
and count, expanded the step card; auto-collapses when done.
* gui: finished plan stays as a Last-plan receipt with a clear button
A superseding plan always replaces it, half-done or not; dismissal is
keyed to plan content so the next plan reappears.
* engine: report window occupancy as a context snapshot
Estimated system/tools/conversation split + resolved window, on ready and
each assistant_message; provider-reported prompt total wins when present.
* gui: right rail glance layer — environment, context meter, header summaries
Environment block (machine/folder/sandbox) atop the rail; Context section with
proportional breakdown and compaction note; collapsed headers carry muted
summary values (amber only for attention); header actions show on hover.
* engine: async subagents — spawn_agent family, human-only approvals, expiry note
Background children (read_only/inherit) with parent-stream approvals and a
fail-closed spawn-time deadline whose tool result says the ask expired;
reports steer or wake the parent; child tokens fold via the agent_task notice.
* gui: Subagents section in the right rail
Active + done runs with seeded icon chips and ages, amber while one waits on
approval; fed by agent_task events + the sessions/tasks endpoint.
* engine: persist subagent runs in a supplemental store
agent_tasks.json beside the wake store, written on transitions: last 50
finished per session, 8KB report cap, live-at-restart runs marked lost.
Also fixes report delivery through call_soon_threadsafe (kwargs crash).
* gui: animal identity glyphs for subagents
Ten one-color silhouettes with punched-out eyes, eight color families per
theme; header chips on the collapsed section; done history compresses to a
glyph grid past ten runs.
* e2e: unshadow the compaction flow; align team/machine specs with the new rail
One compact-the-context handler (signal, divider, context stamp); team header
asserts the attention summary; machine-group locator scoped past the env block.
While GET /v1/providers is on its way the page said "No providers match."
with an empty search box. Now it says Loading providers…, and if the
request fails, that it could not load them, with Try again.
Sessions saved before the model-prefix fix (PR #837) recorded one OpenAI
model as both openai:model-x and model-x, so Usage showed it twice and split
its tokens. Recordings are not rewritten; they are read through one rule:
drop the openai: prefix, keep every other. The server's model_names.py
canonical_model (and its SQL form for the ledger's model grouping) and the
GUI's canonicalModel in modelNames.ts are the same, line for line, as #837's
canonical_model and canonicalModel, so the two agree when both land.
Applied to the Usage grouping by model, each turn's models (Heaviest turns,
How it's been doing), the session usage by model and the Trace panel's
model labels.
Settings ▸ Coworkers and Usage (By coworker, Heaviest turns, the picker,
Ratings) showed the backend's name, "OpenWorker", while the sidebar, top
bar and the coworker page say "Coworker". One helper, listPersonaName, now
names it "Coworker (default)" wherever coworkers are listed side by side;
the coworker page keeps "Coworker". The backend id and name are unchanged.
zh: Coworker(默认).
(a) The history table has fixed columns and wrapping headers, so it stays
inside its card; a version kept without its files is still numbered (v1),
so its row is no longer the long ⓕ fingerprint that wrapped. (b) "too few
sessions to compare" is centred across Typical tokens and Approvals asked.
(d) The coworker's allowed models read "Can use: …" and the history's
"Models used in the last 30 days: …", which now go through the canonical
rule (only the openai: prefix is dropped, as canonical_model on
issue/ope-247). (e) The Default mode and Workspace lines are gone.
(c) and (f) were already fixed in 3942ed9e and 96aae337; the backend on 8765
was started before them and needs a restart to show them.
Versions are v1, v2 (oldest first) with "· current" and the fingerprint
on hover; In use reads "Nov 2" for one day, "… – today" and "Nov 3 or
earlier"; the columns say Sessions, Typical tokens per session (hover: the
middle session's tokens), Approvals asked per session, Turns with issues and
Ratings, and "too few sessions to compare" is one note across both typical
columns. A first version says only "First version seen on this machine.",
has no Show diff ("Nothing to compare yet"), and its models are on their own
line, "Models used: X and N more" with the full list on hover, counting only
the coworker's own work (main and helpers). The default coworker's only
version is marked current again: its current fingerprint is now worked out
from the built agent, as sessions do, since it has no manifest.
"Turn N of M" opens a list of the session's turns: number, start time, the
first words of your message, tokens, an issue badge and the wake reason;
click a line, or type a number and press Enter. A search box keeps the turns
whose message, answer, tool names, file names or issue titles match,
ignoring case, and says "No turns match." when none do. It reads only what
the GUI already holds: the conversation, split at each turn's answer by its
message id, and the recorded turns, which now list their warning checks.
← and → move between turns while the panel has focus and no text box does.
Under the picker, a thin bar per turn: its height is the turn's tokens, amber
when it has issues, the current one marked, click to jump, and a hover line
"Turn N · 12.3k tokens · 1 issue"; past 120 turns the bars are thin and the
strip scrolls.
A short "Active time" header with "Time without waiting" on hover, and air
before Issues, so the two headers no longer run together. The session title
is cut before "· turn N", never the number (the Ratings card too). Model
shows its own name (deepseek-v4-pro) with the full id on hover. Issues is
wide enough for one whole badge. Checked at a 933 px card (18 px text, as in
the user's app) and a 1280 px window, light and dark: only titles are cut,
nothing passes the card's edge.
The period's 👍 and 👎, how often each reason was picked, and the turns
rated 👎 with their note, how many of their steps were auto-approved, and a
Trace link. GET /v1/observability/ratings gives it, for the same turns the
page counts and narrowed to one coworker when one is picked. No charts or
trends; those stay with Langfuse.
In Settings ▸ Coworkers the name opens the coworker's page for every
coworker, OpenWorker included. In Usage ▸ By coworker each name opens that
coworker's page at How it's been doing; Back returns to Activity. A test
checks the default coworker's page and history answer for its id.
#/activity/log and #/activity/usage, so reloading stays on Activity and its
tab (main had no address for it either). With a coworker picked, a line over
By coworker says it lists all coworkers, for comparison.
Fixed column widths; titles, coworkers and long model names are cut with
the full text on hover; one issue badge and +N for the rest, all listed on
hover. No sideways scroll.
Main work is often 97% of the tokens, so summaries, the auto-approver and
session titles were hairlines. Under the main chart, the same days without
main work on their own scale, and a line saying a bar shows each part's
tokens when pointed at.
Each reviewer is named in full (Session model, Decision model · OpenAI
Decisions, Decision model · Jev) with its own checks, overrides, median
time and model tokens, and a Total row carries the shadow note. The server
gives overridden and tokens per reviewer in by_reviewer: an override counts
for the reviewer its turn ran with.
Says which switches, what error texts can contain, and why verdicts are
always sent; the Not sending line says where to pick a destination. Off is
shown as a label, not a pill that looks like a switch.
"Pick any that apply" under the reasons, "Will be saved on this machine"
until Send or Skip, 👎 again closes the card without rating, and 👍 is not
shown chosen while the card is open.
The top-bar button is "Session trace"; the link under each answer and in
the Usage and run-health rows, and the panel title, are "Trace" (zh: 追踪,
会话追踪). Sentences that named Steps say "the trace" now.
Coworker manifests list models as `openai:gpt-5.6-sol`, while the machine's
model list names OpenAI models bare (`gpt-5.6-sol`). The composer compared
the two as plain strings, so on a machine with only an OpenAI key every
such coworker showed "No model" and could not be used. The server made the
same comparison, so a session or team worker could record the model under
either name, which split one model into two in Usage.
One rule now decides model identity: a prefix naming the default provider
is dropped, every other prefix is kept. It lives in
`coworker/providers/router.py` (`canonical_model`) and
`surfaces/gui/src/modelNames.ts` (`canonicalModel`).
- Composer: `coworkerModels` picks the picker list, the unrunnable entries
and the model a fresh draft binds to, using that rule.
- Event-started session settings: same rule for the coworker's model list.
- Server: `resolve_persona_model` and `resolve_worker_model` compare and
return the canonical id, so new sessions and team workers record one
name per model. Existing records keep the name they were saved with.
- The engine's Stop flag is made on the turn's own event loop. A turn on
another loop (each test-client socket has one) could not wait on the
old flag, read the stream as stopped and ended with an empty answer
while its model call ran on outside the turn, so the ledger missed it
and the totals stopped matching the pill.
- The evidence card reads each scanner's own found-something codes.
- The Usage page narrows to one coworker; the turns and approver
endpoints take persona_id.
- A joined machine stamps its turns with the id the controller lists it
under, from the welcome itself.
- The Scheduled view lists joined machines' automations and routes every
call and run session to their machine.
Activity gains a Log | Usage tab: tokens per day by purpose with the
viewer's own days, by coworker, by machine (offline as of), heaviest
turns, the Auto-approve strip and the proxy line. Each answer gets a
footer with time, tokens this turn, issues and Steps, and a rating card.
The coworker page shows How it's been doing per version with What
changed and Show diff. TeamView gains the wake split, team events and
the coverage and evidence blocks. The Scheduled view shows run health,
tokens and issues, and the failed-in-a-row banner. en and zh strings.
Engine: footers carry started_at; team summary items carry their
coverage and findings in the shape the cards read; cap and deferral
tests.
- P3-1: the team rail's token total adds up the lead's and workers'
recordings, so it matches their Steps pills (titles and auto-approver
calls included). Sessions with no recording keep the old count.
- P3-3: the backend closes a manual run when its turn ends, so it no
longer depends on the window. At start, runs an earlier app left
"running" are closed, or marked failed if the app stopped mid-turn.
- P3-4: at start, a background pass works out Issues for finished turns
with no result or one from an older signals.yaml (TD3-8). Results are
stored as seen.
- P3-5: the store's usage_reported column is filled for model steps only.
- P3-6: "sent today" counts from the viewer's midnight, sent by the
desktop as day_start_ms (TD3-2).
- P3-7: a model call with no purpose is "unattributed", not "main".
Every label on the Steps panel, its Usage and Issues tabs explains itself
on hover or keyboard focus (138+ texts, English and Chinese).
Fixes from manual testing:
- the header names a sandbox that enforces nothing ("runner-local · not
enforced") instead of "No sandbox" (F-7);
- the context shows the memory count and mode without ids (F-2), and the
rules and AGENTS.md rows also at 0 (F-5);
- "Audit log: 3 entries for this call: proposed, started, finished"
replaces "Joined records" (F-3);
- "Helpers" and "Helper tokens" tiles, as in the mock (F-8);
- the Issues tab says "19 checks: … found something · … fine · … didn't
apply" and lists what didn't apply with why (F-9);
- issue and fine sentences are built from their facts in the reader's
language; a test checks all 428 golden sentences (C-2);
- "A model call failed" shows the provider's error text (C-3);
- the Observability status counts model calls from outside any session
(F-4).
Each explore row has a summary row the helper's steps fold behind,
"Helper 1 · 6 steps · 12.3k tokens · ok". A lead turn shows each worker
turn it woke as one row with an Open worker session link. English and Chinese.
Turns now record the stretches the machine could not be seen working:
- nobody_watching, on the crossings between no viewer and one;
- controller_offline, from the link's loss until it comes back, saved in the
state folder so a restart in between still writes it;
- clock_jump, when the wall clock runs over 5 s ahead of the turn's clock:
later steps follow the wall clock again;
- recording_gap gains why the turn was found open and when.
Each is a notice with a duration in every running turn, cut at its start and
end. Export sends a gap as its own span.
Signals 0.4: "Waited while nobody was watching" also counts the wait's
overlap with nobody_watching, so a card left in an open session while the
desktop was closed counts (golden S2-17d). The Steps panel draws gaps as
dashed bands with their times, with Machine offline and Nobody watching
totals.
A turn that starts the sandbox records a `sandbox` setup step with its
stages (preparing, the provider's own stages, connect, verify, copy logins)
and which one failed. A restart for changed folders, or a runner that came
back on its own, is a `sandbox_restarted` notice, and the command's
`sandbox.notice` says restart or regrant instead of always null.
The Steps panel shows the setup row with its stages, the failure callout and
the restart row; setup time is its own total. Export sends the setup span
with one event per stage. runner-local now starts on the first turn like
the real providers.
The live export check sent a file's contents with only "Your messages"
on: not through a tool part (those stay out, rule R1), but because the
AI's own answer quoted the file. The switch's help text now says so.
- The page: record steps, keep for 7/30/90 days, save one session's recording; send
recordings to Off, Langfuse or an OpenTelemetry endpoint (address, keys or up to 8
headers, write-only, shown back as masked hints); send a test recording; the five
include switches with the metadata row locked on; the never-sent note; the status card.
- Managed by a company policy: read-only, who set it, the locked destination and
switches, what goes where.
- English and Chinese strings; the #/settings/observability link.
- The in-session notice when a company policy sends content (live and on reload). The
server keeps it on the engine and saves it with the first turn, so an empty draft
session does not appear in the sidebar.
Phase 2, OB-3. Every finished turn is checked against fixed rules (the
shared signals.yaml, 19 checks) and the result is stored with the turn.
The Steps panel gets an Issues tab, and the sidebar an amber ring for
warnings not yet looked at.
- The engine records the recording format 1.2 fields the checks read:
the arguments' hash (RFC 8785), repeat_ok, result_error, error_kind,
reaches_network and the todo plan on tool steps; max_iterations on the
turn; max_output on model steps; inbox_at on waits; whether the
machine expects a sandbox in the stamp.
- coworker/observability/signals.py holds the checks, the same ones that
were validated on the in-app recordings and the shared golden test.
- The checks run when a turn ends; a result made by an older
signals.yaml is worked out again on read. New endpoints serve a
turn's issues and mark a session's issues seen.
- The Issues tab groups issues by check, warnings first, with the step's
command and output line and Go to step; "Checked and fine" is folded.
- tests: the Phase 2 golden test (38 scenarios and hash vectors), the
1.2 fields from a real turn, and the endpoints through the server.
stamp_changed was in the schema but never written. The engine now compares
attendance, the reviewer and the sandbox with the turn's start at each model
call and after the sandbox starts, and writes the notice with only what
changed. The sandbox a first turn makes now reaches the recording, with its
runner version.
The sandbox stamp's site count read a list no provider has, so it was always
empty; it now counts the profile's hosts, the machine's extra hosts and the
hosts the copied logins need.
A lead's steer_worker woke the worker as a channel turn; it is a team wake.
The Steps panel shows what changed when a Settings changed row is opened.
Colours, fonts and layout as the UX-059 mock, light and dark: the step
colour tokens, the time axis with gridlines, and a step's details in a
pane beside the list. The details show a model call's tokens, context,
input and output; a tool's arguments, approval wait and how it was
allowed; a check's questions and the call it judged.
The step endpoint now returns a tool call's arguments, its approval
wait, and for a check the step it judged.
- The Steps button shared the Artifacts button's class, so window-drag.spec's
`.topbar-artifacts-btn` locator found two buttons. It has its own class now,
styled the same.
- test_ws_error_persists_notice_and_retry_reruns: the fire-and-forget title call
could take the one scripted outage meant for the chat call. Title calls are
answered apart, as test_server already does for the concurrency test.
- Header: a pill with the approval mode and, in Auto-approve or Shadow, the
reviewer; a stamp line with machine, sandbox, coworker fingerprint, model
and provider (UX-059 OB-1, OB-12, OB-14, OB-15).
- The Auto-approve line also gives the checks' time and the reviewer.
- A Steps link under each answer opens the panel at that turn.
- Row badges as in the mock: exit code, failed call type, over the context
limit, no usage, auto-approved, shadow checks, Inbox waits, unknown steps.
- Details show what the recording has: request and response ids, attempts,
result size, call id, the context's contents, summary tokens, breaker,
decision-model answers.
- Values in details are words in English and Chinese, not raw codes.
- A session switch opens the newest turn; a narrow rail stacks details.
- Context size shown when the window is not known; a line for steps still
running when the app stopped.
Found by running the panel on real turns:
- picking another turn closes the open step's details (they showed the
old turn's step, with its start measured from the new turn);
- no empty message box under a model call whose reply was only tool calls;
- a duration between 0 and 100 ms reads "<0.1s", not "0.0s";
- a model call's details show the coworker fingerprint, as in OB-2.
Phase 1 screens from UX-059 (OB-1 steps timeline, OB-2 step details,
OB-4 session usage), reading the engine's recording endpoints.
- A Steps button in the session header opens the panel in the right
rail's full-view slot. Turn picker; tiles for Total and every time
bucket that is not zero, plus Other when it is 1% or more; the
auto-approve line; every step of the turn on one time axis, nested
steps (helper calls, a check's model call) under their parent.
- Clicking a row shows its details: status, timing, model and provider,
first reply time, finish reason, tokens split by type (a count not
reported shows as a dash, never 0), the decision path and who decided,
the transcript message, and the exact request and response.
- The Usage tab: input, output and cache share, the at-least and
failed or over-limit lines, by purpose, by model (in, cache, out) and
context now. /usage gains the per-model parts and totals for it.
- The panel refreshes on live step events and at a turn's end.
The Issues tab (OB-3) is Phase 2.