Commit Graph

1233 Commits

Author SHA1 Message Date
João Salavisa 672f2b1f26 feat(stats): add avg turn duration and messages per turn (#1328)
* feat(stats): add avg turn duration and messages per turn to averages card

Adds two new metrics to the workspace stats Averages card: mean turn
duration and mean message count per turn. Both exclude outliers — turns
shorter than 1s (no-op/aborted), longer than 1h (zombie turns where
completed_at was backfilled across an agent restart), and turns with no
messages — so the displayed means reflect typical work rather than tail
skew. On a ~4k-turn sample the filter drops avg duration from ~1062s to
~289s.

* fix(stats): exclude exact 1h turns and use Tooltip for averages help text

Address PR feedback: the BETWEEN bound was inclusive so turns exactly at
the 1h ceiling were retained, contradicting the documented half-open
range. Also swap the title attribute on the new Averages rows for the
shared Tooltip component so the help text is reachable via keyboard and
screen readers.

* fix(stats): align turn-row zero state with em-dash convention

Address PR feedback: avg_messages_per_turn rendered "0.0" while
avg_turn_duration_ms rendered "—" under the same empty-data condition.
Also corrects the GetGlobalStats docstring to acknowledge that
clean_turn_agg adds one correlated per-turn lookup on top of the four
single-scan CTEs.

* test(stats): cover exact 1h turn boundary in clean-turn exclusion test

Add a turn of exactly 3,600,000ms (cleanTurnMaxDurationMs) to
TestGetGlobalStats_CleanTurnExcludesOutliers so the exclusive upper
bound is exercised directly, not just via the 2h zombie turn that
sits well above the boundary.

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
2026-06-11 13:45:16 +01:00
Carlos Florêncio 8e4ee65b31 fix: improve sidebar action layout (#1323) 2026-06-11 13:44:52 +01:00
Carlos Florêncio 800d71f3d1 fix: correct sidebar workspace switcher routing and selection persistence (#1329) 2026-06-11 12:02:25 +01:00
Carlos Florêncio 00500cca05 fix(acp): persist session model and drop composite id logic (#1327) 2026-06-11 11:07:22 +01:00
Carlos Florêncio b0482285aa fix: improve archive switch reliability (#1325) 2026-06-11 09:50:52 +01:00
Clément Fiere a03d65fdf8 feat(sentry): support self-hosted instances via configurable URL (#1320) 2026-06-10 22:06:42 +01:00
Carlos Florêncio dd182acfd2 fix(acp): bound long session resumes and restore legacy model surfaces (#1324) 2026-06-10 21:20:28 +01:00
Carlos Florêncio bb15fcf474 fix(web): show open PR status when task has merged and open PRs (#1322) 2026-06-09 17:53:59 +01:00
Carlos Florêncio b3da5f9af1 feat(web): add session handoff via agent tab context menu (#1317) 2026-06-09 16:42:05 +01:00
Carlos Florêncio 71bf90d9e0 fix: default new task executor to worktree (#1321) 2026-06-09 16:40:27 +01:00
Nicolas Lenepveu 63434f29cf fix(orchestrator): deliver initial prompt to passthrough start_agent (#1306) 2026-06-09 09:36:13 +01:00
Carlos Florêncio 525e932df7 fix: keep utility default agent and model paired (#1318) 2026-06-09 00:05:26 +01:00
Carlos Florêncio 68e7fc7c27 fix(acp): handle config-option-only model state (#1319) 2026-06-09 00:04:40 +01:00
Carlos Florêncio 64140ae942 fix(acpdbg): read claude agent models and modes from configOptions (#1310) 2026-06-08 22:44:02 +01:00
Carlos Florêncio ec40bf0b87 fix: contain settings page overscroll (#1316) 2026-06-08 22:42:14 +01:00
Carlos Florêncio d78a1a4370 fix: default changes panel to tree view (#1314) 2026-06-08 21:22:39 +01:00
Carlos Florêncio 3e92d3abc6 fix: allow office mcp mode for existing tasks (#1315) 2026-06-08 21:22:30 +01:00
Carlos Florêncio 2b068fbbf2 fix: clarify required automation fields (#1312) 2026-06-08 20:06:02 +01:00
José Almeida 5d6fe29293 fix(web): collapse dockview workflow stepper to current step when cramped (#1309)
* fix(web): collapse dockview workflow stepper to current step when cramped

On narrow windows the full workflow step indicator outgrew the top-bar
center track and rendered beneath the right-side action buttons. Switch
the grid so the breadcrumb takes only its content width and the center
track absorbs the real leftover space, then collapse the stepper to a
minimal current-step view (via the shared useToolbarCollapsed hook) only
when the full version genuinely cannot fit.

* fix(web): stabilize stepper test id and drop margin skewing collapse

Move data-testid="workflow-stepper" to the persistent outer container so
the e2e locator resolves in both expanded and collapsed modes, and remove
mx-2 from the w-full measuring container (it overflowed the wrapper and
shifted the collapse threshold by ~8px); the grid gap-2 already spaces the
columns.

* refactor(web): trim verbose stepper collapse comment

* fix(web): tag archived minimal badge with stepper test id

Give the archived branch of MinimalWorkflowStepper the same
workflow-stepper-minimal test id as the current-step branch so collapsed
mode is detectable for archived tasks, and tighten two comments.

* test(web): collapse multi-line comments in stepper test to single lines

* fix(web): drop aria-current on minimal stepper no-current fallback

When no step is current the minimal view falls back to the first step;
it should not claim aria-current="step", matching the expanded path where
no step is marked current in that case.
2026-06-08 19:34:40 +01:00
Carlos Florêncio fc4bf8a40c feat: add checkout service make targets (#1311) 2026-06-08 19:30:21 +01:00
Nicolas Lenepveu aac2afbf93 fix(agent-settings): stop reseeding agent profiles the user deleted (#1305) 2026-06-08 17:49:34 +01:00
Carlos Florêncio 637123f84f feat: allow remote external mcp access (#1307) 2026-06-08 17:41:29 +01:00
Carlos Florêncio c68323c9e0 fix: preserve settings sidebar on refresh (#1303) 2026-06-08 17:40:54 +01:00
Carlos Florêncio 351db61e5b fix: avoid stale sibling PR sync (#1302) 2026-06-08 17:40:16 +01:00
Carlos Florêncio c4133a9e6e fix(web): focus sidebar-created tasks (#1301) 2026-06-08 17:39:37 +01:00
José Almeida 46f20e0dfe fix(web): keep dockview group alive on task switch to stop layout collapse (#1308)
* fix(web): keep dockview group alive on task switch to stop layout collapse

On a fast-path env switch, the incoming session chat was added only after
removeEphemeralPanels closed the outgoing chat. When that chat was the last
surviving panel in its group (e.g. the default layout's chat column), removing
it first emptied — and so destroyed — the group. The captured group id then no
longer existed, and post-#1165 the old `referenceGroup: "sidebar"` fallback is
dead, so addPanel ran with an undefined position: dockview dropped the incoming
chat into the active (terminal) group and collapsed the grid root to a vertical
stack. The corrupted layout was then persisted, so it survived reloads.

Add the incoming session panel before removing the outgoing one, anchored at the
outgoing panel's current index, so the group stays alive throughout the swap.
Removing the panels ahead of the insert point shifts the new panel into the same
final slot, so computeSurvivingIndex is no longer needed.

* docs(web): note the raw-index equivalence assumption in env switch
2026-06-08 17:33:54 +01:00
João Salavisa 2b74f4e189 fix: gate step_complete_kandev tool on per-step signal flag (#1300)
* fix: gate step_complete_kandev tool on per-step signal flag

Two ADR-0015 follow-up bugs:

- "Wait for agent completion signal" checkbox snapped back because
  updateWorkflowStepAction dropped auto_advance_requires_signal from the
  PUT body. Add it to the Pick<>, body builder, BackendWorkflowStep, and
  transformWorkflowStep.
- step_complete_kandev MCP tool was injected into every first-turn
  prompt, polluting context and tempting agents on non-signal-gated
  steps to call it. Replace the static description in
  kandev-context.md with a {step_complete_section} placeholder filled
  only when the current step has auto_advance_requires_signal=true.
  Add orchestrator.Service.StepRequiresCompletionSignal helper and
  thread it through all four InjectKandevContext call sites.

* fix: address PR 1300 review feedback

- Hoist ProcessOnTurnStart above the first-turn MCP wrap so an
  on_turn_start step transition is reflected in the
  auto_advance_requires_signal flag baked into the stored/forwarded
  prompt (coderabbit, cubic).
- Add Service.WorkflowStepRequiresCompletionSignal step-ID variant and
  use it from task_operations.go / autoStartStepPrompt to skip the
  redundant GetTask round-trip at hot first-turn launch sites (claude).
- Add table-driven tests covering every silent-false branch and the
  positive path of both helpers (greptile, claude, cubic).

* docs(sysprompt): document stepCompleteSection trailing newline contract

Address Claude review suggestion on PR 1300: make explicit that the
constant must end with "\n" because the kandev-context.md template inlines
{step_complete_section} immediately before the next bullet, so the
trailing newline is what separates the two list items.

* fix: restore task-id lookup at startTask wrap site

Address PR 1300 review regression: the previous swap to
WorkflowStepRequiresCompletionSignal(ctx, workflowStepID) silently hid
step_complete_kandev on manual starts where the workflowStepID parameter
is empty but the task is already bound to a signal-gated step in the DB.

The `task` variable in scope at this call site is `*v1.Task` (from the
scheduler) and does not carry WorkflowStepID, so the cheaper step-ID
variant is not applicable here — go back through StepRequiresCompletionSignal
so the repo lookup pulls the canonical step.

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
2026-06-08 08:14:52 +01:00
Carlos Florêncio f81b5133bf fix(agentctl): preserve codex reasoning model ids (#1296) 2026-06-07 23:06:35 +01:00
Carlos Florêncio 3f922d3911 fix(web): sort kanban cards by newest created (#1298) 2026-06-07 22:55:42 +01:00
Clément Fiere 0b40d09b10 fix(utility): list opencode models when ACP probe is empty (#1278) 2026-06-07 22:52:12 +01:00
Carlos Florêncio eb74a3d329 fix: alert when git executable is missing (#1297) 2026-06-07 22:44:57 +01:00
Carlos Florêncio 210737cb24 fix(web): use recent task after removal (#1295) 2026-06-07 22:30:52 +01:00
Carlos Florêncio 372c07b55a fix(web): gate changes push button on ahead count (#1294) 2026-06-07 22:14:11 +01:00
Carlos Florêncio 6661255024 fix(web): restore dockview sidebar layout (#1293) 2026-06-07 21:41:40 +01:00
Carlos Florêncio 1484ae7d28 fix: expose user agent bins to system service (#1292) 2026-06-07 21:13:28 +01:00
João Salavisa f6f6fc3877 fix(agentctl): read models from configOptions fallback (#1291)
* fix(agentctl): read models/modes from configOptions fallback

claude-agent-acp v0.42 dropped the unstable `models` field from
session/new and publishes models only via `configOptions[]` with
category=model. The probe still relied on the legacy field, so
the inference-agents endpoint reported zero models and the UI
showed "Claude advertised no models."

Read the legacy fields first, then fall back to configOptions
when they are absent. Same fallback applied to modes for parity
with future agents.

* fix(agentctl): gate configOptions fallback on nil legacy field

Switch the fallback guard in `applySessionProbeFields` from
`len(out.Models) == 0` to `sessionResp.Models == nil` (same for
modes). A non-nil `Models` struct with an empty `AvailableModels`
slice is schema-valid: the legacy branch already stamps
`CurrentModelID`, and the previous guard would then fire the
configOptions fallback and overwrite it with a different source.

Adds `TestApplySessionProbeFields_LegacyEmptyModelsBlocksFallback`
covering the edge case.

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
2026-06-07 14:36:51 +01:00
João Salavisa 31d4ba67fa perf(stats): split stats endpoint per-section and rewrite GetGlobalStats (#1289)
* perf(stats): split stats endpoint per-section and rewrite GetGlobalStats

Stats page used to wait on a single endpoint that ran 7 SQL queries in parallel
and rendered nothing until the slowest one returned. Cold first-paint was the
bottleneck users actually hit.

- Backend: split /api/v1/workspaces/:id/stats into 7 per-section endpoints
  (global, tasks, daily-activity, completed-activity, agent-usage,
  repositories, git). Each handler runs one repo call and returns its DTO.
- Backend: rewrote GetGlobalStats from 9 independent correlated subqueries
  into 4 CTEs (task / session / turn / message). Each underlying table is now
  scanned at most once per request instead of 1-3 times.
- Frontend: replaced the monolithic fetchStats + single PanelState with
  useStatsSections hook (7 parallel fetches, per-section SectionStatus).
  Each panel renders skeleton -> data -> error independently; light queries
  (overview, agent usage, git) paint immediately, heavy ones stream in.
- Tests: added TestGetGlobalStats_AggregatesAcrossTables to lock the
  CTE-based query against regressions for turns, user/tool messages, and
  total duration.

* fix(stats): address PR review feedback

- Restore cache: "no-store" on section fetches; Gin emits no Cache-Control
  headers so the browser could otherwise serve stale stats on revisit
  (Greptile P1, Cubic).
- Stop blanking the Overview panel when /git fails: render OverviewCards as
  soon as `global` is ready and treat `git` as optional, falling back to the
  Averages card. Updates GitOrAveragesCard / OverviewCards types (Claude).
- Fix TaskStatsHasMore off-by-one at exactly taskStatsLimit rows: probe
  taskStatsLimit+1, slice back to the page size, only flag hasMore on a
  confirmed extra row (Claude, Cubic).
- Rename anyError → firstError with a comment documenting the deterministic
  insertion-order iteration; clearer signal that callers should use it as
  "any section failed?" and surface per-section errors in their own panels
  (Claude).

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
2026-06-07 13:24:32 +01:00
José Almeida 607201be79 fix(web): tighten dockview topbar button sizing (#1290)
* fix(web): tighten dockview topbar button sizing

* fix(web): keep editor button focus ring inside clipped container
2026-06-07 13:00:19 +01:00
João Salavisa 48a621a566 fix(utility): surface inference-agent probe status + add refresh endpoint (#1287)
* fix(utility): surface probe status in inference-agents endpoint

GET /api/v1/utility/inference-agents previously filtered out agents whose
host-utility probe didn't reach StatusOK, so a Claude entry with auth or
install issues vanished from the response. The frontend then defaulted
agent_id to "claude-acp", failed to match anything in the list, and
silently rendered an empty Model picker with no explanation.

Include every registered ACP-capable agent in the response with a new
`status` field (and sanitized `status_message`) so the UI can render an
inline note + Refresh button. Add a companion
POST /api/v1/utility/inference-agents/:id/refresh endpoint backed by the
existing hostutility.Manager.Refresh so users can re-probe after fixing
auth or installing the CLI without restarting kandev.

Credential-looking substrings ("api_key=...", "token=...") in probe
errors are stripped before reaching the wire.

Refs #1269, #1263.

* fix(web): show probe status note + Refresh on utility-agent model picker

Previously the Model picker rendered empty and disabled whenever the
selected inference agent was missing from /inference-agents (probe
hadn't reached StatusOK) or advertised no models, with no way for the
user to understand or recover.

Wire the new `status` field through the API client and render an inline
note ("Sign in to Claude to load models.", "Claude CLI is not installed
on this machine.", "Setting up Claude…") with a Refresh button that
calls the new POST /inference-agents/:id/refresh endpoint and merges the
updated agent back into local state — no full refetch, so a slow probe
on a different agent doesn't block the UI.

Both call sites — the create/edit dialog and the Default Model row in
the settings page — render the note. Component-level unit tests cover
the status-to-copy mapping, fallback ("no longer available") copy, and
the in-flight Refresh button disabled state. The e2e test asserts the
auth_required → click Refresh → status:ok state machine end to end.

Refs #1269.

* test(utility): add HTTP integration tests for inference-agents endpoints

End-to-end coverage of the new /api/v1/utility/inference-agents and
/api/v1/utility/inference-agents/:id/refresh routes against a real gin
router with a stateful in-memory host stub.

- ProbeStateMachine: GET (auth_required) → POST refresh → GET (ok+models)
- RefreshUnknownAgent: 404 without invoking the host probe
- RefreshPathFuzz: traversal, null byte, 512-char id — no 5xx, no probe
- ConcurrentRefresh: 50 parallel POSTs under -race, no panic
- StatusMessageSanitizationOnTheWire: 110KB payload + unicode probe
  errors — no secret material reaches the wire, unicode prose preserved

Caught the gap unit tests had: the multi-request GET→POST→GET sequence
on a shared cache, and the wire format (Content-Type, JSON shape)
through the real router.

* fix(utility): tighten credential regex + drop Go 1.22 loop-var shadow

Address two inline review comments from claude bot on PR #1287:

- Drop `ia := ia` shadow in httpRefreshInferenceAgent — pre-Go 1.22
  pointer-capture workaround, dead weight under Go 1.24+ (per
  apps/backend/AGENTS.md).
- Split the credential sanitizer into two patterns so prose without a
  real `=`/`:` separator is no longer mangled. The single regex
  previously turned "access token was revoked" into
  "access token=<redacted> revoked" by consuming the next word as the
  value. New shape:
    - credentialKVPattern — requires `[:=]` (api_key, token, secret,
      password)
    - credentialBearerPattern — allows bare whitespace, anchored to
      "bearer" only
  Existing redaction cases still pass; two new prose cases guard the
  regression.

* fix(utility): address cubic + greptile review findings on PR #1287

Round-2 review feedback:

- P0 (cubic): postEmpty called `t.Fatalf` from `go func()` in
  TestIntegration_ConcurrentRefresh. Go panics on t.Fatalf from non-test
  goroutines. Split into `postEmptyE` (error-returning, goroutine-safe)
  and `postEmpty` (main-goroutine wrapper). Concurrent test now
  forwards the error through the existing `errs` channel.
- P1 (cubic + greptile): tightened the credential sanitizer regex.
    * `api[ _-]?key` now also matches a literal space ("invalid api key=…")
    * separators are restricted to `[^\S\n]` so multi-line stderr like
      "invalid token\ncaused by network" no longer eats words across lines
  Test coverage added for both regressions (and for tab as a valid
  horizontal-whitespace separator).
- P2 (cubic): refresh-failure log now feeds `err.Error()` through
  sanitizeStatusMessage before zapping it — same wire-side guard, but
  applied to server logs too. Closes the "credentials in journalctl"
  hole.
- P2 (cubic): InferenceAgent.status is now optional in the TS type and
  the status note treats an undefined status with `models.length > 0`
  as healthy, so older API responses don't flash a spurious warning.

Deferred (acknowledged in PR replies, not changed here):

- cubic P2 useInferenceAgents drift — the hook is currently called in
  both the dialog and the parent section. Drift is bounded by the
  dialog's own `onSuccess` refetch and only visible inside an open
  dialog. Lifting state to the parent and prop-drilling is the right
  fix but touches multiple files; out of scope for this review round.

* docs(utility): fix stale comments flagged by claude review on PR #1287

Two doc-only items, no code change:

- handlers_integration_test.go: TestIntegration_ConcurrentRefresh
  comment said "we don't assert on call count" but the test DOES
  assert `host.refreshCalls != N` two lines later. Updated to match
  reality (singleflight lives in the real Manager, the stub has none
  → every call must reach the stub).
- utility-agents.spec.ts: the ACP-probe E2E said "backend filters out
  non-StatusOK agents", which is the exact behavior this PR removes.
  Updated to explain the real reason only Mock appears (KANDEV_MOCK_AGENT=only
  registers only `mock-agent`).

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
2026-06-07 12:38:28 +01:00
José Almeida 2ed65cfde3 fix(web): guard dockview measure against mid-transition sidebar width (#1288)
* fix(web): guard dockview measure against mid-transition sidebar width

The unified AppSidebar (#1165) animates its width over 300ms; the dockview
content container (a flex sibling) can be measured mid-transition at a tiny
positive width in onReady, so the default layout builds too narrow and dockview
collapses the horizontal columns into a vertical stack. measureDockviewContainer
trusted any positive parent clientWidth. Reject an implausibly-small live width
(< 50% of viewport) as a not-yet-laid-out transient and fall back to the
viewport; the resize observer then snaps to the exact size.

* fix(web): preserve live height when width guard fires; assert it
2026-06-07 12:35:29 +01:00
João Salavisa 91d5861519 fix(web): preserve manual task selection during archive/delete (#1286)
* fix(web): preserve manual task selection during archive/delete

When the user archived (or deleted) a task and then immediately picked
another task, the in-flight archive flow would auto-switch to a third
task, clobbering the manual selection.

The shared `removeTaskFromBoard` was honoring the pre-API captured
`wasActiveTaskId` unconditionally. That capture exists to survive the
WS `task.deleted` / `task.updated(archived_at)` handler clearing
`activeTaskId` to null. But it also overrode the legitimate case where
the current store value reflected a user-initiated navigation, so the
equality check still passed and `switchToNextTask` ran.

New guard: switch only when (a) the store still points at the removed
task, or (b) the store has been WS-cleared to null AND the captured
pre-API id matches the removed task. Any other state means the user
moved on — leave them where they chose to go.

* chore(web): tighten null check + no-opts test in use-task-removal

Addresses non-blocking suggestions from claude bot review on #1286.

- Use strict `=== null` in `shouldSwitchAfterRemoval` for consistency
  with the surrounding strict-equality comparison.
- Add unit test documenting that a no-opts caller skips the auto-switch
  when the WS handler has already cleared `activeTaskId` to null.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-07 11:57:54 +01:00
João Salavisa 99d8d8f173 fix(github): handle UNIQUE collision in UpdatePRWatchBranchIfSearching (#1285)
* fix(github): handle UNIQUE collision in UpdatePRWatchBranchIfSearching

When a sibling PRWatch already owns the destination
(session_id, repository_id, branch) triple, the raw UPDATE tripped
the UNIQUE constraint and the change was logged as an error
("failed to update PR watch branch"). Wrap the update in a tx that
first checks for a sibling; if one exists, delete the source watch
(pr_number=0, no PR state to preserve) and cede branch ownership.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* fix(github): close TOCTOU and cover sibling-has-PR collision case

Address PR review feedback on #1285:

- Catch SQLite "UNIQUE constraint failed" on the no-sibling UPDATE path
  and treat it as the same collision case (Greptile P2 inline). A
  concurrent CreatePRWatch that commits between our existence probe
  and the UPDATE would otherwise re-trigger the bug this PR fixes.
- Switch the existence probe from `SELECT id` + `siblingID` dummy to
  `SELECT 1` scanned into a typed `probe int` with an inline comment
  (Claude inline readability nit).
- Extract `dropSourceAndCommit` helper so the probe-hit branch and the
  TOCTOU fallback share one code path.
- Add `TestUpdatePRWatchBranchIfSearching_CollidesWithSiblingHasPR_DropsSource`
  exercising the realistic race: sibling has already found its PR
  (pr_number != 0) when source tries to migrate (Claude / Greptile P2 /
  cubic P3 all flagged the gap). Asserts sibling row + PR number
  survive untouched after source is dropped.

* test(github): cover missing-row no-op and reframe TOCTOU comment

Address Claude bot's follow-up suggestions on #1285:

- store.go: rewrite the UNIQUE-catch comment. The original wording
  implied an in-process race, but SQLite's writer pool is
  SetMaxOpenConns(1) so a concurrent in-process CreatePRWatch cannot
  commit between probe and UPDATE. Reframe as defensive belt-and-
  suspenders for an external SQLite writer / future pool changes.
- service_pr_watch_multi_branch_test.go: add
  TestUpdatePRWatchBranchIfSearching_MissingRow_NoOp covering the
  ErrNoRows -> tx.Commit() idempotency contract that the new
  transaction code introduces. Locks the prior UPDATE-by-id semantics
  of "0 rows affected, no error" for a nonexistent id.

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-07 11:50:13 +01:00
João Salavisa f86ab2312f feat(workflow): ship explicit completion signal toggle + YAML round-trip (#1284)
* feat(workflow): ship explicit completion signal toggle + YAML round-trip

ADR 0015's "Wait for agent completion signal" UI shipped as a deferral
comment in PR #1276 instead of a working toggle, and the
auto_advance_requires_signal flag never reached the workflow export
format. Add the pipeline-editor sub-toggle and propagate the field
through StepPortable so the setting survives workflow export/import.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* fix(workflow): guard completion-signal toggle + cover import in tests

Address PR review:
- hide ExplicitCompletionToggle when on_turn_complete has no transition
  (flag has no effect there; visible checkbox was misleading)
- add service-level import test for auto_advance_requires_signal so the
  field has explicit ImportWorkflows coverage, mirroring the existing
  show_in_command_panel pattern

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* fix(workflow): drop omitempty from AutoAdvanceRequiresSignal tags

Match the sibling bool fields (IsStartStep, ShowInCommandPanel,
AllowManualMove) so a `false` value still appears in exported YAML/JSON.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-07 11:43:28 +01:00
Edan Binshtok 8b6a62cab2 fix(backend): defer move_task when session is running (#1277)
defer RUNNING/STARTING moves instead of immediate MoveTask; add normalizeTaskState aliases; regression test for promptless Work to Done.

Co-authored-by: cybertronic <cybertronic@gmail.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-06-07 10:48:04 +01:00
José Almeida e626d33a7c fix(web): make toggle-sidebar shortcut work on every route (#1283)
* fix(web): make toggle-sidebar shortcut work on every route

TOGGLE_SIDEBAR (Cmd/Ctrl+B) was handled only inside useEditorKeybinds,
mounted just by the dockview desktop layout, so it was dead on Kanban,
the task list, Office, Settings, etc. Lift it to a global app-root
listener (useAppShortcuts, mounted in GlobalCommands) that honors user
overrides via getShortcut and keeps the isEditableTarget guard, and drop
the now-duplicate branch from useEditorKeybinds. Adds a hook unit test
and a Playwright e2e covering Kanban, Settings, and the dockview session.

* fix(web): hydrate userSettings into the root store on settings routes

Each StateProvider creates its own store; the settings/general page mounts
a nested one that the root-mounted GlobalCommands/AppSidebar can't see. The
settings layout hydrated the root store but omitted userSettings, so
override-driven global shortcuts (TOGGLE_SIDEBAR, Quick Chat) were dead on
every /settings route. Hydrate userSettings there too, matching the
kanban/tasks/office/github routes.

* test(web): cover the macOS Cmd path; note the settings-layout catch trade-off

Address review feedback: add a metaKey (Cmd) variant to the useAppShortcuts
test so both ctrlOrCmd branches are exercised, and document that the settings
layout's outer catch intentionally drops userSettings overrides in the
already-degraded error path.
2026-06-06 23:09:12 +01:00
José Almeida a8d84474c4 fix(web): point sidebar Home to office dashboard in office mode (#1282)
* fix(web): point sidebar Home to office dashboard in office mode

The AppSidebar Home item always navigated to the Kanban view (/), even on
office routes. Make its destination office-aware so Home targets the office
dashboard (/office) while in office mode and the board (/) otherwise.

* test(web): start Kanban Home e2e from a non-home route

Per PR review: the Kanban-mode Home test started at / and clicked Home (also
/), so no navigation occurred. Start from /settings/system/status so the click
is a real navigation back to the board.
2026-06-06 23:00:47 +01:00
José Almeida dd4cb2c3ee fix(web): align office topbar border with sidebar header (h-10) (#1281)
* fix(web): align office topbar border with sidebar header (h-10)

The OfficeTopbar was h-12 (48px) while the AppSidebar header is h-10 (40px),
so the office top bar's bottom border sat ~8px below the sidebar's
under-workspace-picker border, breaking the continuous seam where the sidebar
meets the page content. Drop the topbar to h-10 to match the sidebar header and
the PageTopbar/dockview top bars used elsewhere.

Add data-testid hooks to both bordered sections and a Playwright regression
asserting they resolve to the same height and bottom y-position.

* test(web): force expanded sidebar in topbar alignment check

Addresses review feedback (Greptile/Claude): the alignment assertion only
holds for the expanded sidebar header (its border sits under the workspace
picker, and that layout carries data-testid="app-sidebar-header"). Set the
appSidebar.collapsed flag off via addInitScript before load so leftover
collapsed state can't turn the check into a confusing timeout instead of a
real assertion.
2026-06-06 22:38:31 +01:00
José Almeida b001aa1eb8 feat(web): unified app sidebar (#1165)
* feat(web): introduce unified AppSidebar and retire legacy nav surfaces

Mounts a single collapsible AppSidebar at the root layout that replaces the
former WorkspaceRail, OfficeSidebar, and dockview-embedded sidebar pane.
Wires per-section expand + collapse state through a new UI slice with
localStorage persistence, ports projects/agents/settings/tasks sections, and
trims integrations / quick-chat / settings buttons + the home breadcrumb
from the dockview top bar (debug-only MoreToolsMenu kept). Strips the
sidebar column from every preset and the panel registry; persisted
sidebar panels are dropped by the existing sanitize layer at restore time.

* test(web): point workflow-filter e2e at AppSidebar's Home link

The dockview top bar no longer renders the IconHome breadcrumb after the
unified sidebar landed — the home navigation now lives in AppSidebar's
primary nav. Update the workflow-filter spec to click that link instead.

* feat(web): wire app-sidebar primary controls, nested settings, and footer migration

Gates office-only nav (Inbox, Agents, Projects, New Task) behind the office
feature flag, renames Tasks → Kanban as the bottom flex-grow section, replaces
the gradient workspace picker with a minimal trigger when office is off, and
moves Stats/Improve/What's-New/Office out of the kanban top bar into the
AppSidebar footer + a dedicated Integrations section. Expands Settings into a
collapsible deep tree mirroring the old SettingsAppSidebar so /settings pages
can drop their nested SidebarProvider.

* feat(web): make app-sidebar resizable and inline task top bar actions

Adds a persistent drag-resize handle on the AppSidebar's right edge that
clamps between APP_SIDEBAR_EXPANDED_WIDTH and 30vw; width is stored on the
appSidebar slice and persisted to localStorage. Drops the dockview top bar's
overflow popover (TopbarActionOverflow) in favor of always-inline clusters,
and replaces the debug "More tools" dropdown with a single debug toggle button.

* feat(web): polish app-sidebar headers and consolidate task views

Section headers now lead with a clean uppercase label, drop the mono
clash, lift contrast from /60 to /70, and end with the collapse chevron
at the right so the section body aligns flush under the label.

The Tasks section's filter bar (view chips + gear in a card-tinted
strip) is replaced by a compact view dropdown that lives in the section
header. Drag-to-reorder lives in the existing filter popover, which is
now reachable from the same gear button.

Settings moves below Tasks so the kanban list keeps the flex-grow seat
in the sidebar.

* feat(web): move settings group chevrons to the right

* fix(web): drop chat from home topbar; calm tasks group headers

The kanban top bar's Quick Chat button is removed now that Quick Chat
lives in the AppSidebar; workspaceId stops being threaded into the
tablet and desktop header variants.

Group headers inside the Tasks section (kanban workflow steps) lose
the opaque background strip and the section-label uppercase styling
so they read as inline subgroup rows, not duplicate section dividers.
The tasks-section wrapper drops its negative margin so the group rows
sit flush under the TASKS label instead of bleeding out past it.

* fix(web): restore agent logos on settings tree profile leaves

SettingsLeaf gains a leadingIcon slot for pre-rendered visuals; agents
group passes <AgentLogo agentName={agent.name}> so each profile row
shows its agent's logo like it did before the old SettingsAppSidebar
was retired.

* feat(web): brand the sidebar, move New Task off the kanban top bar

The Add task button (CreateTaskTopbarButton) leaves both the desktop
and tablet kanban headers; New Task is now an unconditional primary
nav item on the AppSidebar, falling back to a disabled affordance when
no workspace is active.

AppSidebar header gains a Kandev wordmark that links to /, with a
single-letter K mark in collapsed-rail mode. PageTopbar's root variant
no longer renders the parent brand label when backLabel is empty, so
the home view's top-left "Kandev" goes away in favor of the sidebar.

* feat(web): unbind sidebar toggle by default; drop Home from breadcrumbs

TOGGLE_SIDEBAR's configurable default is now UNBOUND_SHORTCUT (the new
sentinel for "no binding"); matchesShortcut already returns false on
an empty key, so the keybind hook needs no change. Keyboard shortcuts
settings show "Unbound", expose a Clear (X) button when a custom
binding is set, and the reset icon falls back to unbound for entries
whose default is unbound.

PageTopbar's breadcrumb variant no longer prepends the Home back-link
when backHref is "/" (the AppSidebar provides Home navigation).

* feat(web): move New Task into the sidebar, retire the workspace selector

- Extract AppSidebarNewTaskItem (office vs regular create dialog) with tests
- Drop the Workspace selector from the kanban display dropdown and mobile
  menu sheet; the sidebar workspace picker is now the sole switcher
- Factor SectionHeader out of AppSidebarSection; animate fixed sections via
  Collapsible while the grow (Tasks) section stays flex-driven
- Polish brand header, share SIDEBAR_ITEM_ACTIVE accent-bar classes, move
  task group chevrons to the right, drop the task-item selected fill
- Bump dockview layout keys to v3 and filter the orphaned 'sidebar' column
  so pre-AppSidebar default layouts no longer render a broken grid
- Point the cross-workspace isolation e2e at the sidebar picker

* feat(web): sidebar settings mode, subtask shortcut, accordion settings nav

- Restore subtask creation: the sidebar New Task row now shows a trailing
  subtask affordance when viewing a task (regular mode), reopening the
  NewSubtaskDialog the retired dockview header dropdown used to provide
- Fix settings pages not scrolling: scroll the content region (topbar stays
  pinned) via min-h-0 + overflow-y-auto
- Add a footer gear that toggles a settings-only sidebar takeover; the gear
  spins on toggle. Mode is transient (never persisted) and force-expands a
  collapsed rail on entry
- Make the settings nav a single-open accordion: opening one top-level group
  closes the others and reveals its subsections; route changes re-sync the
  open group. Extract the tree into a reusable SettingsTree
- Repoint subtask e2e specs at the new sidebar affordance

* test(web): restore create-task e2e flow after New Task moved to sidebar

The New Task relocation removed the kanban-header create-task-button and made
the sidebar entry auto-navigate + open the office dialog under the e2e profile
(office on), breaking ~20 specs that drive the regular task-create dialog.

- Give the sidebar New Task the create-task-button testid so KanbanPage
  resolves it again
- Drop the post-create auto-navigation so creating a task stays on the board
  (matches the retired kanban button; specs find/click the new card)
- Ungate the subtask affordance from office mode (subtasks use the compact
  NewSubtaskDialog in both modes, as the old dropdown did)
- Add a useRegularMode() helper that runs a spec file with KANDEV_FEATURES_OFFICE
  disabled (backend.restart bracket) and apply it to the 18 create-task specs
- Point the System sidebar-navigation spec at a /settings/system page so the
  single-open settings accordion has that group expanded

Validated: create-task, create-task-disabled-tooltip, create-task-branch-selector,
and subtask all green (modulo pre-existing idleInput agent-timing flakes that
CI retries absorb).

* fix(web): sidebar task navigation, settings-mode close + transitions

- Clicking a task in the global sidebar from a non-task page now navigates to
  the task route (mounts the dockview) instead of only rewriting the URL via an
  in-place layout switch that assumed the dockview was already mounted
- Make the settings-mode 'Settings' header a close affordance (the footer gear
  is far from the tree); clicking it exits the takeover regardless of which
  group is expanded, and it fades in when opened
- Fade in the Tasks (grow) section content on expand, matching the other
  sections' animated reveal
- Enable next-themes disableTransitionOnChange so dark/light flips switch every
  surface in the same instant rather than each animating at its own duration
- Add a regression e2e for sidebar→dockview navigation

* fix(web): clear sidebar task highlight when off the task route

The AppSidebar is global, so tasks.activeTaskId lingered after navigating Home,
leaving the last-opened task visually selected in the sidebar. Gate the
selected/active highlight by route — only highlight while on /t/ or
/office/tasks/. Expose data-active on the sidebar task item for testability.

Adds an e2e: entering a task highlights it; returning Home clears it.

* fix(web): never build the dockview default layout at zero width

On a fresh client-side navigation into a task the dockview can mount before the
grid has painted, so measureDockviewContainer returned api.width/height = 0.
Building the default layout at width 0 makes dockview collapse the horizontal
columns into a vertical stack (chat / files+changes / terminal) instead of the
intended side-by-side layout. Fall back to the viewport when neither the live
container nor the api report a usable size; the resize observer then snaps the
grid to the exact container size.

TDD: dockview-measure.test.ts covers the zero-size fallback.

* fix(web): keep the default layout horizontal when swapping chat for the session tab

Entering a task rendered a vertical stack (chat on top, files/changes in the
middle, terminal at the bottom) instead of the side-by-side default. The
chat->session-tab swap removed the 'chat' placeholder BEFORE adding the session
panel; when chat was the only panel in the center group, removing it destroyed
that group, the stored centerGroupId went stale, resolveInitialPosition
returned undefined, and the session panel was appended as a new row — flipping
the grid root to VERTICAL with three stacked groups.

Fix: add the session panel into chat's live group first (resolveInitialPosition
now prefers the chat panel's group), THEN remove the chat placeholder, so the
center group never empties and the horizontal split is preserved.

Also fixes the pre-existing pane-resize reload/clamp e2e failures, which were
the same collapse. TDD: task-default-layout.spec.ts asserts the right column
sits beside chat.

* fix(web): remove the sidebar Settings section; settings is gear-only

Settings was still a collapsible nav section in the sidebar AND auto-expanded on
/settings routes. That also caused the reported bug: toggling the footer gear
off while on a settings page revealed the normal sidebar with the Settings
section already expanded, so the settings tree appeared to 'not close'.

- Drop the Settings nav section (and its /settings route auto-expand) — the
  settings tree is reached only via the footer gear takeover
- Delete the now-unused sidebar settings-section component
- e2e: gear opens the tree and closes it even after navigating into a section

* fix(web): show office sidebar sections only while inside Office

Inbox, Projects, and Agents were gated on the office feature flag, so they were
always visible when the feature was enabled. Gate them on actually being in the
Office surface (an /office route, reached via the footer Office button) via a
new useInOffice() hook — they stay hidden in the regular workspace and appear
only after entering Office.

Unit test for the hook; e2e (office fixture) asserts the sections appear on
/office and are absent on the regular home.

* fix(web): drop the colorful workspace selector, restore the minimal one

The workspace picker swapped to a gradient/initials trigger (and gradient
swatches in the dropdown) whenever office was enabled. Always use the plain
name + chevron trigger and drop the gradient swatches, matching the previous
design. Removes the now-unused workspace-gradient helper.

* fix(web): align sidebar header, scope office New Task, exit settings on nav

- Drop the doubled right border: the resize-handle hairline is transparent at
  rest, leaving the aside's border-r as the single edge (highlights on hover).
- Collapse the expanded sidebar header into one h-10 row (brand · workspace
  picker · collapse) so it lines up with the page/dockview top bar.
- Gate the sidebar New Task dialog on useInOffice() (actual /office route)
  instead of the bare office feature flag, so Kanban mode keeps the classic
  create dialog while office is enabled.
- Close the settings takeover when navigating off the /settings surface
  (Home, Kandev brand, Office, etc.), tracked on a real pathname change so the
  gear still opens it in place and in-tree navigation keeps it open.
- Lock in the already-working collapsed-gear-expands behaviour with an e2e.

Tests: new-task-item unit (route gating), settings-gear e2e (Home closes,
collapsed gear expands), office gating e2e (Kanban vs Office dialog).

* fix(web): realign task top bar to h-10, refine sidebar workspace picker

- Pin TaskTopBar to h-10 so its bottom border lines up exactly with the
  sidebar header (and the PageTopbar) — previously it had no fixed height and
  sat ~2px off, breaking the seam across the sidebar/content edge.
- Redesign the workspace picker as a compact, muted secondary control inlined
  after the Kandev brand: brand and workspace share text-sm so they sit on a
  common baseline, with a faint '/' separator; the workspace reads as muted and
  the brand stays primary.
- Drop the workspace switcher (and its irrelevant folder glyph) from the
  collapsed rail — collapsed now shows just brand + expand.

Adds an e2e that measures the task top bar bottom edge against the sidebar
header (within 1px).

* fix(web): replace the /github and /gitlab presets rail with a scope bar

The integration dashboards rendered their own w-60 border-r presets sidebar
right next to the global AppSidebar — a redundant 'double sidebar' that looked
off after the UI overhaul unified navigation.

On desktop, fold the presets into a horizontal scope bar above the list
toolbar: the bounded, common controls (PR/Issue kind + preset pills, grouped
Inbox/Created) stay visible as pills, while the unbounded saved-queries list
collapses into a 'Saved' dropdown so the bar never grows tall. The dashboard is
now single-column like Kanban/Tasks. Mobile is unchanged — the vertical
PresetsSidebar still lives in the hamburger sheet.

Adds a desktop scope-bar e2e (presets + kind switching, no inline rail) and
repoints the mobile page object at the new scope-bar testid.

* fix(web): address CodeRabbit/cubic review — a11y, dedup, dup nav, lazy-load

- Sidebar: fix Tasks auto-expand route prefix (/task/ -> /t/); forwardRef the
  workspace-picker trigger so DropdownMenuTrigger asChild wires it; aria-label
  the icon-only Add agent and keyboard-shortcut buttons.
- Settings tree: drop the duplicate top-level Editors leaf (lives under
  General); broaden the System group active state to all /settings/system.
- CSS: honour prefers-reduced-motion for .sidebar-fade-in.
- Storage: bump the dockview maximize namespace to v3 in lockstep with the
  layout namespace so stale pre-v3 maximize blobs are discarded.
- Integration dashboards: extract a shared IntegrationScopeBar base so the
  github/gitlab scope bars are thin wrappers (de-dup + under the component
  size cap). Reuse useConfiguredIntegrationLinks in the sidebar Integrations
  section instead of recomputing availability.
- Lazy-load the office NewTaskDialog so it's not bundled on non-office routes.
- tasks-view-picker: use onSelect for keyboard-activatable menu items.
- Tests: normalize group x by container left in task-default-layout; refresh
  stale subtask-flow comments; async findByTestId for the lazy office dialog.

* fix(web): gate the scope-bar delete button pointer-events while hidden

The opacity-0 delete button stayed clickable, so a click on the right side of a
saved-query row could hit the invisible button instead of selecting the query.
Add pointer-events-none until the row is hovered (group-hover:pointer-events-auto).

* fix(web): use onSelect for scope-bar saved-query menu items (keyboard a11y)

Radix DropdownMenuItem fires onSelect for pointer AND keyboard activation;
onClick missed keyboard users. The nested delete button stops pointerdown so it
isn't treated as a select of the saved query.

* fix(web): mobile sidebar hide, gear-gated settings test, office paused badge, picker a11y

- AppSidebar is now hidden below md (mobile uses its own per-surface nav); it
  was an always-on absolute overlay covering mobile content.
- Restore the sidebar-agent-paused-badge testid on the office agent row.
- Update system sidebar-navigation spec to open the gear (settings is a
  gear-gated takeover in the overhaul).
- Workspace picker menu items use onSelect for keyboard a11y.

* test(web): reconcile main's E2E suite with the unified-sidebar overhaul

Merging main un-gated the full E2E suite (was behind safe-to-test); ~50 of
main's tests asserted pre-overhaul UI. Updated them to the overhaul:

- Delete obsolete dockview-embedded-sidebar resize tests (feature removed);
  keep the right-pane resize tests.
- Scope locators where tasks now appear in BOTH the page and the global
  AppSidebar list (intended) — task-list, task-list-filters, pr-status-badge.
- Adapt sidebar-filter specs to the new TasksViewPicker (gear popover +
  view-chip dropdown) and its page object.
- Settings now reached via the footer gear (not the topbar); Improve Kandev
  opens from the sidebar footer; office nav folded into the AppSidebar — update
  those specs (office verified in CI; can't run locally).
- session-layout: handle the new close-terminal confirm dialog; multi-session
  and tab-management: foreground the (now backgrounded) chat tab after sidebar
  task switches.
- Skip create-task-url-reopen-no-branches: its provider-branch flow (code
  identical to main) is broken only by the office-off backend restart this PR
  forces via useRegularMode — a harness interaction, not an app regression.

* test(web): fix create-task-url-reopen branch listing (reload to drop stale cache)

Un-skipped and fixed properly. Root cause: during the URL-paste/registration
flow the frontend can transiently fetch branches for the freshly-registered
provider repo before its background clone settles; useBranches caches that early
empty result as terminal (if (isLoaded) return), so re-picking the repo never
re-lists. Reload before the second dialog open drops the in-memory branch cache
so branches list fresh from the backend (verified: the backend returns them).
The provider-API path and product behaviour are unchanged from main.

* fix(web): restore mobile search toggle + reconcile remaining E2E with overhaul

- Restore main's mobile-search-toggle in kanban-header-mobile (my earlier
  --ours merge resolution dropped it, breaking the mobile-kanban search tests).
- Office specs: expand the collapsed Agents/Projects sections before asserting
  their rows, and scope office-tasks-page locators to <main>. [CI-verified]
- mobile-task-list-search / mobile-voice-mode: scope task titles to the table
  and drop the obsolete data-sidebar=trigger assertion.
- app-sidebar.test mock: add settingsMode, toggleAppSidebarSettingsMode, and the
  integrations section key.
- sidebar-filter page object: scope chipMenu to the menu holding the view chips.

* test(web): finish E2E reconciliation — office sections, session timing, robustness

- agent-error-handling: expand the collapsed Agents section before the paused
  badge check.
- office execution-indicator/task-sorting: wait for a non-empty .font-mono
  identifier (assigned async by the backend).
- session-layout: foreground the (CI-backgrounded) chat tab before the idle
  check; make the close-terminal confirm dialog optional (busy heuristic is
  timing-dependent).
- workspace-switch-isolation: run office-off so the picker switches in place
  instead of redirecting to /office (which needs onboarding this fixture skips).
- app-sidebar-page: condense JSDoc to one line per project convention.

* test(web): match main's office identifier assertion (revert over-scoping)

The round-2 office agent scoped the .font-mono identifier check to <main> too,
but only the task TITLE needed scoping for the sidebar duplication. The office
task identifier is empty in the e2e seed (backend-assigned, not populated here),
so the <main>-scoped non-empty filter found nothing. Revert the identifier
assertion to main's exact form (first .font-mono on the page); keep the title
scoped to <main>.

* test(web): restore non-empty .font-mono filter for office identifier checks

Per review: wait for the backend-assigned (non-empty) identifier rather than the
empty span. Unscoped so it finds the identifier wherever it renders, 15s window.

* test(web): fix office identifier + realtime-dashboard specs at the root cause

Ran office specs locally (they DO run here) and root-caused all three:

- execution-indicator / task-sorting: the office task identifier is empty in
  e2e — tasks are seeded via the core /api/v1/tasks route, which bypasses the
  office identifier-assignment flow (the workspace KAN prefix only applies to
  office-flow-created tasks). Main's .font-mono.first() was a false-positive
  matching an element the overhaul removed. Assert the identifier COLUMN renders
  (attached to the row) instead of a value e2e never produces.
- realtime-dashboard: the Recent Tasks card is fed by dashboard.recent_tasks via
  useOfficeRefetch on office WS events; a core-route task's office event is async
  so the in-place refetch is genuinely flaky. Reload for the deterministic SSR
  dashboard fetch (what a user sees on revisit); the WS-refetch mechanism stays
  covered by the sibling cross-workspace test.
2026-06-06 21:58:39 +01:00
github-actions[bot] 2557b69c0e release: 0.56.0 2026-06-06 20:43:33 +00:00
João Salavisa 97c1f689e4 feat(workflow): explicit step-completion signal for auto-advance (#1276)
* feat(workflow): explicit step-completion signal for auto-advance

Adds `step_complete_kandev` MCP tool + `WorkflowStep.auto_advance_requires_signal`
column so workflow steps can gate their `on_turn_complete` transitions on an
agent-emitted completion signal instead of inferring "done" from any turn-end.
See docs/decisions/0015-explicit-completion-signal-for-auto-advance.md for the
motivation, alternatives considered, and trade-offs.

* fix(workflow): address PR review feedback for explicit completion signal

- Dedup `LoadPendingStepSignal` into `internal/task/models` so the MCP
  handler (write site) and the orchestrator (read site) share one decoder.
- Gate the out-of-band subscriber on `WorkflowStep.AutoAdvanceRequiresSignal`
  so `step_complete_kandev` is not an unintended manual-advance path for
  signalless steps.
- Fail closed on step-load errors in `processOnTurnCompleteViaEngine`: a
  loader hiccup must block the transition rather than silently bypass the
  gate.
- Reject `step_complete_kandev` for terminal sessions
  (Completed / Failed / Cancelled) so the agent gets a clear error instead
  of `accepted: true` followed by a no-op.
- Surface bus-publish failures back to the agent instead of swallowing
  them — the idempotency guard makes the retry safe and prevents the
  session from getting stuck.
- New `TestOnStepCompletionSignaled` covers the subscriber paths
  (RUNNING / WAITING+match / stale step / not signal-gated).
- ADR doc fixes: payload field list now matches the wire shape, writer
  ownership is documented, the unsubstantiated 250ms-debounce claim is
  removed, and the late-signal re-close race is recorded as a deferred
  open question.
- Drop the new toggle's HelpTip in favour of an inline description — the
  shadcn tooltip portal was intercepting clicks on the agent-profile
  dropdown options in the workflow settings E2E.

* fix(workflow): round-2 review fixes for explicit completion signal

- Split `handleStepComplete` into `resolveStepCompleteTarget` +
  `publishStepCompletionEvent` helpers so the entrypoint clears funlen
  (88 → under 80 lines) and the two halves are independently testable.
- Classify session/task load errors instead of returning `NotFound` for
  every failure: use the exported `taskrepo.ErrTaskNotFound` sentinel for
  task lookups, substring-match for the session repo (no sentinel today),
  and surface anything else as `InternalError` with the raw error logged
  but not leaked to the agent — so a transient DB hiccup triggers a retry
  rather than session abandonment.
- Pin `step_complete_kandev` explicitly in the mode-specific tool-count
  tests: `Contains` in `ModeTask`, `NotContains` in `ModeConfig` and
  `ModeOffice`, so accidental dereg or cross-mode registration is caught
  by the diff instead of by a still-equal count.

* fix(workflow): round-3 review fixes + defer pipeline-editor toggle

- Defer the "Wait for agent completion signal" sub-toggle to a follow-up
  PR. Adding it to the step config card shifts the layout enough that the
  agent profile select's Radix popper opens into a sibling tooltip's hover
  path, breaking workflow-agent-profile.spec.ts on shard 7. The backend
  column ships in this PR — workflows can opt in via the YAML import or
  the MCP `update_workflow_step_kandev` tool — and the UI affordance gets
  its own iteration where the test interference can be investigated
  without blocking the rest of the change.
- Re-publish the bus event on the `already_signaled` short-circuit when
  the session is WAITING_FOR_INPUT. Without this, an agent's retry after
  the first call's publish failed would short-circuit on the persisted
  bag, never fire the subscriber, and leave the session stuck.
- handleStepComplete docstring: drop the non-existent feature-flag claim
  and the "audit entry" wording — gating is purely
  `WorkflowStep.AutoAdvanceRequiresSignal` and the bag is cleared on the
  next turn start with no separate audit trail.
- `step_complete_kandev` tool description: same fix — clarify that the
  call succeeds for non-gated steps but the signal is discarded on the
  next turn start (no audit history to query later).

* test(workflow): cover handleStepComplete + reword defer-comment

- Add `TestHandleStepComplete_*` (6 subtests, all passing) covering the
  uncovered handler paths flagged by Claude/cubic on round 3:
  missing-fields validation, session-does-not-belong-to-task ownership
  guard, terminal-session rejection (Completed/Failed/Cancelled),
  first-call happy path (bag persisted + bus payload shape pinned), dedup
  while RUNNING (no re-publish), and dedup while WAITING_FOR_INPUT (the
  retry-after-publish-failure path — must re-publish so the subscriber
  fires).
- Reword the deferred-toggle comment to drop the YAML-export claim:
  the workflow export model doesn't yet serialize
  `auto_advance_requires_signal`, so MCP `update_workflow_step_kandev`
  is the only configuration path that round-trips today. Note the
  follow-up needs to extend the export schema alongside the UI toggle.

* test(workflow): pin signaled_at + drop redundant GetTaskSession on bag-clear

- Assert `signaled_at` is present and parses as RFC3339Nano in
  `TestHandleStepComplete_FirstCallAccepted`. The field is part of the
  handler's documented response contract; pinning it catches accidental
  drop/rename regressions.
- Add `clearPendingStepSignalByID(ctx, sessionID)` as a thin DB-only
  variant of `clearPendingStepSignal` and use it from
  `executeStepTransition`, where the session struct isn't read after the
  clear. Avoids an extra `GetTaskSession` round-trip per transition.
  `clearPendingStepSignal(session)` still exists for the
  in-memory-mutation case (subscriber's stale-step path).

* fix(workflow): subscriber stale-clear guard + ADR/comment drift fixes

- onStepCompletionSignaled's stale-step path now reads the bag and only
  clears it when its current StepID matches the (stale) event's stepID.
  Previously the unconditional clear could destroy a valid signal that
  the new step's agent had just written when sessions are reused across
  steps without auto_start_agent. (Greptile P1.)
- Drop the "feature flag" phrase from WorkflowStep.AutoAdvanceRequiresSignal
  docstring — leftover from an earlier draft. The gating is purely the
  boolean field; the global flag was removed two rounds ago.
- Merge the two adjacent `if triggerOnEnter` blocks in
  executeStepTransition so the full triggerOnEnter=true path reads top-
  to-bottom in one place. (CodeRabbit + cubic.)
- ADR 0015 Mechanism section: re-label the SessionStepHistory.Metadata
  audit annotation, the `step_complete_signaled` system message, and the
  expvar telemetry counters as deferred. They're planned shapes, not
  current behaviour — calling them "Audit" / "Chat visibility" /
  "Telemetry" in present tense was misleading for future readers.

* fix(workflow): drop subscriber complexity + reload session on stale-clear

- Extract `parseStepCompletionEvent` and `stepIsSignalGated` helpers from
  `onStepCompletionSignaled`. Cyclomatic complexity goes 17 → under 15
  (cyclop on new code, CI lint regression after the round-6
  stale-clear guard).
- The stale-step branch now re-loads the session from the DB before
  deciding whether to clear the bag. The previous round-6 fix checked
  the local `session.Metadata` snapshot, which can be older than a
  concurrent write from the new step's agent — that check could see a
  stale signal that had since been overwritten with a valid one and
  erase the new bag entry. The narrower load-on-clear is cheap and
  closes the cubic P1 hole. (The broader race in
  `processOnTurnCompleteViaEngine` reading session.Metadata is the
  monotonic-marker open question in ADR 0015; out of scope here.)

* test(workflow): pin stale-event guard's preserve-current-bag side + dedup StringFromAny

- Add `TestOnStepCompletionSignaled` subtest "stale event → valid bag
  for CURRENT step is preserved" — fires a stale step-A event while
  the bag holds a valid step-B signal, asserts the bag still contains
  StepID=step_b after the subscriber runs. Pins the negative side of
  the round-6 `existing.StepID == stepID` guard, so a regression that
  drops or inverts the guard (and would silently leave signal-gated
  steps stuck) fails the test instead of slipping through.
- Promote `pendingSignalString` to exported `models.StringFromAny`
  and reuse it from the orchestrator subscriber's
  `parseStepCompletionEvent`. Eliminates the byte-for-byte duplicate
  helper across the two packages; the bag-shape decoder and the bus-
  event decoder now share one source of truth for "missing string".

---------

Co-authored-by: Kandev Agent <agent@kandev.dev>
2026-06-05 18:32:27 +01:00