Der Queue-Picker gatete eine task_dependencies-Kante auf status='done'.
review_task(skipVerify) laesst einen erfolgreich gemergten Task aber
bewusst auf WaitingForReview stehen, bis verify_merges einmal zentral
laeuft -- damit blockierte jeder Nachfolger ueber den gesamten Batch plus
Full-Suite-Verify hinweg, bei freien Slots.
Eine Kante gilt jetzt als erfuellt, wenn der Vorgaenger 'done' ist ODER
sein Merge gelandet ist (worktrees.state = 'merged'). TaskBlocking
(get_task/list_tasks/wait_for_task_change) und der UI-Wartehinweis folgen
derselben Regel, damit "blocked" nicht meldet, was der Picker claimt.
Abgeleitet: revert_merge/abort_merge setzen den Worktree auf Active und
blockieren wieder; reject_rerun/reject_park haben nie gemergt und bleiben
unerfuellt; ein Task ohne Worktree wird nur ueber 'done' erfuellt; ein
fehlgeschlagener verify_merges gibt Nachfolger frei, weil der Code so
oder so auf main liegt.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Measured 2026-09-16 against the live endpoint: a clean burst allows exactly 5
requests before a 429, with Retry-After: 300 (a flat 5-req/5-min quota, no
ratelimit-* headers on either the 200s or the 429s). Lowered
UsagePollIntervalActiveSeconds/UsagePollIntervalIdleSeconds default from
300/900 to 120 each (clamp floor 90, up from 60) — leaves ~2.5 requests of
headroom per window for manual refreshes and any other consumer of the same
OAuth token, unlike 60s which consumes the whole quota by itself. The quota
is a flat per-account limit, not tied to ClaudeDo's own running/idle state,
so both keys share the same default now (kept separate for config
compatibility).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
TaskRowViewModel had its own ParseDiffStat expecting "+12"/"-3" tokens that
`git diff --stat` never emits, so the row badge always showed 0/0. Meanwhile
DetailsIslandViewModel's parser matched only the English words
"insertion"/"deletion", so it broke on a localized git (e.g. German Windows:
"Einfügung(en)"/"Löschung(en)").
Row and detail badge now share one parser, anchored on git's untranslated
(+)/(-) suffix instead of the language-dependent word before it, so it reads
correctly in any locale. Also corrected the Worker CLAUDE.md's branch-name
claim (`claudedo/{taskId[:8]}`) to match the actual full-GUID branch name.
A standalone parent (PlanningPhase.None) with existing Idle/Queued children
now runs itself first, then sequences its children instead of flat-enqueuing
all of them at once: SubmitForChildrenAsync calls
PlanningChainCoordinator.SetupChainAsync(enqueue: true) right after the
Running -> WaitingForChildren transition, mirroring the planning path but
firing immediately instead of waiting for an explicit "Queue plan". Planning
parents are left untouched (they queue via QueuePlanAsync), and a child
already Running makes SetupChainAsync throw, which is now caught and logged
so it can't fail the parent's own transition.
Group C of the operation-feedback batch is fully merged; the two consumers
(DetailsIslandViewModel.ApproveReviewAsync, MergeModalViewModel) now handle
the verifying phase from the generic OperationProgressEvent handler instead
of a separate forwarder event.
ConnectedClientTracker zaehlt verbundene UI-Clients (WorkerHub.OnConnectedAsync/
OnDisconnectedAsync). StartListHandler prueft ihn vor dem Broadcast und liefert
bei 0 verbundenen Clients Requested=false plus Reason statt unbedingtem true.
Tool-Beschreibung und external-mcp.md an das reale Verhalten angepasst.
Ein Task kann jetzt auf beliebig viele Vorgaenger warten. Die Einzelspalte
tasks.depends_on_task_id wird durch die Beziehungstabelle task_dependencies
ersetzt (Composite-PK, beide FKs ON DELETE CASCADE -- das reproduziert das
alte ON DELETE SET NULL: geloeschter Vorgaenger = Kante weg = Task wieder
pickbar). BlockedByTaskId bleibt unberuehrt.
- Migration AddTaskDependencies kopiert jede bestehende depends_on_task_id
verlustfrei in die Tabelle, bevor die Spalte fallengelassen wird
- QueuePicker gatet per doppelt negiertem NOT EXISTS: kein Task mit 0 Kanten
wird blockiert, alle Kanten muessen done sein, eine Kante auf einen
verschwundenen Task blockiert weiter
- SetDependsOnAsync ersetzt die ganze Menge; Zykluspruefung ist jetzt eine
BFS ueber den Graphen, faengt also auch A->B->C->A
- blockedReason nennt alle offenen Vorgaenger (External/TaskBlocking, von
ExternalMcpService und TaskWaitMcpTools gemeinsam genutzt)
- add_task/update_task nehmen dependsOnTaskIds; die alte Skalarform bleibt
als Ein-Element-Menge inkl. ""-Clear-Sentinel gueltig
- UI: Fan-in-Zeile bleibt flach und listet alle Vorgaenger im After-Label,
die lineare Kette rendert unveraendert
The status PATCH needs statusName alongside statusId or the API 400s,
so every write-back had been silently failing. The board's flat
assigneeName is null on every live ticket; the real assignee list is
ticketAssignees[]. Import now also picks up In Bearbeitung tickets,
not just Offen ones.
Both issues are live-verified against the real API, so the feature
flag flips back on.
ScopeGlobs auf Tasks und SerializeOnFileOverlap auf der Listen-Config waren
ungenutzt: der Scope wurde nie befuellt, also hat der Queue-Picker nie
serialisiert. ScopeOverlap, das Picker-Gate, die DTO-Felder, die UI-Option und
die Spalten fallen weg (Migration DropFileScopeSerialization).
Tickets/-Ordner, TaskStateService.NotifyAsync, ticket_api_base_url und TicketRef/
TicketProjectId dokumentiert. Verbatim-Copy-Warnung um TicketProjectId ergänzt und
den PermissionMode-Vorbestandsbug (SetConfigAsync liess das Feld im Update-Zweig
aus, gefixt in f0a3a186) vermerkt. docs/open.md um die fuenf offenen Punkte aus der
manuellen Verifikation des Specs ergänzt, inkl. Hinweis dass die Bandel-Response-
Shapes nie live verifiziert wurden.
WorkerHub and IWorkerClient each carried their own copy of every record and
had already drifted — MergePreviewDto lost its verify fields on the client
side, two records disagreed on their name. They now live in Data/Wire.cs and
reach both sides via a <Using> item, so a hub signature change is a compile
error instead of a silently dropped JSON field.
Set Stateless = true on the internal (/mcp on the SignalR port) and external
MCP HTTP transports. Two effects:
- No Mcp-Session-Id, so a worker restart can no longer 404 a session that
outlives it -- ConPTY tiles in the UI process and externally registered
claude sessions keep working across a restart.
- A 2026-07-28 client is no longer refused back to the initialize handshake.
A stateful server rejects that revision on purpose (see the SDK's
StreamableHttpHandler.s_sessionSupportingProtocolVersions), so the new
per-request protocol path was unreachable regardless of SDK version.
Nothing here used the stateful-only features (sampling, elicitation, resource
subscriptions, unsolicited notifications). In-tool progress notifications ride
the POST's own response stream and are unaffected -- which matters, since ~20
tools use them to hold off the client's 300s idle abort. All three consumers
(TaskRunner, PlanningSessionManager, the installer's RegisterMcpStep) already
register type: "http", so dropping the legacy SSE endpoint breaks nothing.
Verified against the running worker across two restarts: tools/list returns
57 tools over a bare POST with no initialize and no session id, and a full
2026-07-28 tools/call round-trip returns real data.
MarkDoneAsync was Idle-only, so a finished task (typically a list-handler
run with no worktree to merge) could not be ticked off. Guard on Running
instead and cover the other statuses with tests.
The task-list done toggle (both islands) and RemoveFromQueue wrote TaskEntity.Status
directly via EF, bypassing TaskStateService: no TaskUpdated broadcast, no guard against
a concurrent picker claim (lost update), and no status-based filter. Added guarded
MarkDoneAsync/UnmarkDoneAsync/DequeueToIdleAsync transitions plus matching hub methods
(SetTaskDone/UnsetTaskDone/DequeueTask) and IWorkerClient wrappers; the three UI call
sites now route through the hub with optimistic-then-revert row updates and
ErrorReported on failure. RemoveFromQueueAsync dequeues each queued child individually
through the same guarded path instead of cascading via a raw EF update.
Also closes two hub guard gaps: UpdateListConfig's delete branch now preserves a list's
SerializeOnFileOverlap flag instead of dropping it, and SubmitTaskForReview's Idle/Failed
status gate now runs before either mutation branch so a Done/Cancelled task can't get
committed or stamped and then rejected.
A child merge that came back blocked/verify_failed/untracked_collision during a
parent/children unit merge used to vanish: DrainAsync only logged it server-side,
PlanningMergeAborted carried no reason, and ApproveReview/review_task always
reported StatusMerged for a task with children regardless of the real outcome,
so a failed unit merge left the parent stuck with no visible error.
- PlanningMergeOrchestrator.StartAsync/ContinueAsync/DrainAsync now return a
PlanningMergeResult(Status, Reason) instead of void, and PlanningMergeAborted
carries that reason to the UI.
- WorkerHub.ApproveReview and ExternalMcpService.ReviewTask's approve branch
propagate the real status/reason for a parent with children instead of
hardcoding "merged" (or masking a non-conflict failure as "conflict").
- StartAsync now requires the parent to already be WaitingForReview for
improvement parents too, not just planning ones, so a stale caller can no
longer trigger a partial child merge.
- HasActiveMerge now also covers the window between the last child merging and
FinalizeParentDoneAsync completing, closing a gap where a concurrent Cancel
could race the parent's own approve-to-Done transition.
- IslandsShellViewModel.OnPlanningMergeAborted flashes the reason via
FlashFooterError instead of only clearing the external-merge banner.
continue_merge and the parent/children unit-merge drain (PlanningMergeOrchestrator)
re-run the post-merge verify gate but never forwarded their IProgress token into it,
so a slow verify command on either path went silent past Claude Code's 300s MCP
idle-abort even though D1-D3 already fixed this for merge_task/review_task's
childless path. list_worktrees also gets elapsed-time progress: many tracked
worktrees means many concurrent git subprocess spawns.
Worker CLAUDE.md's existing progress rule now points at ProgressReporter as the
one implementation instead of a fresh polling loop.
Merge/verify phases now broadcast over a generic (opKey, phase, current,
total) Hub event instead of a merge-specific one, so future producers
(worktree cleanup, startup recovery, planning integration) can reuse it.
IWorkerClient.MergeProgressEvent stays as a thin forwarder for existing
consumers (elapsed seconds riding in the generic "current" slot).
The merge commit message was hand-rolled per caller ("Merge task: <title>",
"Merge <branch>", "Merge subtask") and ignored the task's commit type. Every
caller now passes a blank message and TaskMergeService fills in
CommitMessageBuilder.BuildMerge -> {commitType}(list-slug): merge <title> plus the
ClaudeDo-Task trailer; the merge modal prefills it from GetMergeTargets.
A merge whose list has a verify command holds the MergeTask call for minutes (5m46s
on this repo), during which the modal only disabled its button - no spinner, no
message, so a landed merge looked like a dead app. TaskMergeService now broadcasts
MergeProgress(taskId, phase, elapsedSeconds) for the merging and verifying phases
(re-reported every 30s) plus a WorkerLog line when verify starts; the modal shows a
spinner and the localized phase.
Document task-number allocation in Data layer (TaskEntity.Number,
next_task_number counter, invariants, insert paths).
Update Worker docs to clarify TaskIdResolver wiring (#123 → GUID lookup),
Number in MCP payloads, and correct the 'Two hard conventions' statement
(only the first is test-enforced).
Bump external-mcp.md verified-against commit to 38af549 (Slice 4 merge)
and add new sections on task ID resolution and numbering.
Add Slice 4 visual verification items to open.md (row/detail number display,
worker-log messages).
Verified against:
- src/ClaudeDo.Data/TaskNumberAllocator.cs
- src/ClaudeDo.Data/Repositories/TaskRepository.cs (AddAsync line 20, CreateChildAsync line 276)
- src/ClaudeDo.Worker/External/TaskIdResolver.cs
- src/ClaudeDo.Worker/External/ExternalMcpService.cs (TaskDto/TaskRefDto DTOs)
- tests/ClaudeDo.Worker.Tests/External/ExternalMcpToolSchemaTests.cs (only first convention test-enforced)
SetListConfig previously could only pass VerifyCommand through unchanged;
only the UI hub could set it. Add an optional verifyCommand parameter with
matching clear/merge/create-vs-delete handling, and split the list result
into ListConfigDto so the task-only SetTaskConfig/GetTaskConfig DTOs stay
untouched. Also corrects Worker/CLAUDE.md's claim that tasks can override
verify_command — it's list-only.
Adds a user/MCP-declared task dependency (DependsOnTaskId) distinct from the
planning chain's internal BlockedByTaskId: add_task/update_task can set it,
the queue picker skips a Queued task until the dependency reaches Done, a
Failed/Cancelled dependency leaves the dependent blocked instead of starving
silently, and setting a link rejects self-reference/unknown-id/cycles.
get_task/list_tasks/batch_get_tasks now report blocked/blockedReason, and
wait_for_task_change reports "Blocked" immediately instead of running out its
timeout on a task the picker will never claim.
preview_merge/preview_merge_set gain staleFiles: files a branch touches that
the target branch also changed since the branch's fork point, a more honest
staleness signal than `behind` alone.
Every tool description now leads with what the tool does AND when to reach for
it, since MCP clients rank tools by that text. Per-parameter prose moved onto
the parameters as [Description], exhaustive result-shape enumerations and
design/history rationale dropped, and the repeated boilerplate clauses
(lean-task-ref, batch cap, refused-while-Running) pulled into McpToolDocs,
which also documents the style for future tools.
Tool-level description text: 20494 -> 13605 chars (-34%); combined with the new
parameter descriptions 18517 (-10%).
Closes gaps that caused wrong calls rather than just verbose ones:
- list_task_attachments returns metadata only, no file content
- run_task_now shares continue_task's single override slot and throws when busy
- list_runs is ordered oldest-first and feeds get_run
- workingDir on create_list/update_list is an existing local git repo path,
unvalidated until the first task run
- get_task_worktree's behind=0 also means the main ref was unreachable
Removes get_task_status_values: a whole tool entry for static reference text.
GetTask's description is now the canonical place for status meanings.
Documents the Checks/ + SystemCheckPage feature (gating rule, check list,
FreshInstall-only placement) and the ExecutableResolver/.cmd-shim wiring in
ClaudeDo.Worker's ClaudeCliPreflight/ClaudeProcess. Adds docs/explore-notes/installer-preflight.md
(bumped/corrected against the actual implementation) and links it from the
explore-notes README and root CLAUDE.md.
The underlying code lives only on unmerged task branches (06aca9b3.../40272c0b...),
not on main yet, and two follow-up features ("Claude Help Me" button, Config-mode
Diagnose section) were never implemented because they blocked on that same missing
merge. Both gaps are called out explicitly in the new docs and in docs/open.md's
verification checklist, instead of being asserted as done.
Two remaining single-line lists that accumulate entries over time
(table names in Data/CLAUDE.md, HubBroadcaster event names in
Worker/CLAUDE.md) collide when parallel tasks each append a new
entry, same as the AppSettingsEntity/migration-list lines fixed
earlier today. Content unchanged, one bullet per existing item.
EnsureExists blindly seeded ~/.todo-app/prompts/*.md with the bundled
default and never revisited it, so any file created by opening the
Files settings tab shadowed every later default change permanently
(SuggestImprovement/AskUser sections never reached real runs since
2026-06-04). PromptFiles now hashes what a file was seeded/saved with
in prompts/.defaults.json: Classify() tells missing/current-default/
known-past-default/edited apart, ReconcileStaleDefaults() drops files
that only ever matched a superseded default, and real edits are left
alone and surfaced in the Files tab with a diff + reset-to-default
action. QuarantineOrphans() moves stale-named leftovers (agent.md,
planning.md) into prompts/_orphans instead of silently deleting them.
Wired as a Worker startup sweep (PromptFileRecovery) alongside the
existing OrphanRecovery/AttachmentOrphanRecovery services.
## Kontext: Limits sind Fenster, nicht Summen
Die Runs laufen ueber das Claude-Abo. Limits greifen pro 5h-Fenster und pro 7 Tage. Nicht die Wochensumme tut weh, sondern dass ein Agent-Burst ein Fenster leerraeumt, in dem Mika selbst interaktiv arbeiten will.
## Messgrundlage (alle Transcripts unter ~/.claude/projects)
Agent-Runs sind ueber die ganze Historie nur **18,4 %** des Account-Verbrauch
ClaudeDo-Task: 87105f5e-c4f4-4af4-ae60-89cd2e153e3c
task_runs.tokens_in only ever held the API's uncached "input_tokens" field
(off by a factor of ~400,000 on a resumed session), and tokens_out summed
only the last result event instead of the whole session. TaskRunner now
reads each run's cache-read/cache-write/input/output totals from the
session transcript via a new ITranscriptUsageReader.ReadSessionTotalsAsync,
storing the delta against prior runs on the same session so a --resume
doesn't double-count. New task_runs.cache_read_tokens/cache_write_tokens
columns; the Session tab now shows the raw total (what actually counts
against the 5h/7d usage limit) with a breakdown tooltip.
Runaway sessions were the single biggest cost driver: model_presets was
never persisted (stayed code-only), default_max_turns shipped at 100, and
ResolveMaxTurns had no upper bound, so a task/list override could run
hundreds of turns unchecked.
- TaskRunner.ResolveMaxTurns now clamps the resolved value to
AppSettings.MaxTurnsCeiling (new column, default 80) and logs a warning
with task id / requested / effective value when it clamps.
- default_max_turns default lowered from 100 to 40 (entity, EF config,
and the seeded row via the new AddMaxTurnsCeiling migration).
- AppSettingsRepository.GetAsync backfills model_presets with the
shipping defaults on first read instead of leaving the column null.
- Settings > General's per-model preset table and the task/list agent
editor now show a hint when a set max-turns value exceeds the ceiling.
Add a dedicated Usage Monitor & Gate section to Worker CLAUDE.md, wire
UsagePillViewModel/UsageMonitorModalViewModel and the new IWorkerClient
usage surface into Ui CLAUDE.md, and record open verification points
for the pill/modal visual pass and the real gate E2E in docs/open.md.
Data CLAUDE.md already covered the new entity columns and migration.
Adds GetUsageSnapshot/GetModelUsage/GetTaskUsage to WorkerHub (backed by a
shared UsageSnapshotBuilder), a UsageUpdated broadcast fired after every
UsageMonitorService poll cycle, and records the resolved model on each
task_runs row so per-model/per-task usage can be reported from history.
Persists the merge commit SHA on WorktreeEntity for every successful
single-task and unit merge, and adds a TaskMergeService.RevertMergeAsync
+ revert_merge MCP tool that undoes a merged task's merge via
`git revert -m 1` (never reset/rewrite, since the target checkout is
shared). Rejects cleanly when there's no recorded SHA, the repo is
mid-merge/mid-revert, or the target has foreign uncommitted changes;
a conflicting revert aborts immediately. Also exposes the new
mergeCommit field via get_task_worktree.
Planning sessions could already steer a subtask's model but not its turn
budget, so a visibly large subtask would still die at the global default
turn limit. maxTurns is optional (default null = inherit list/global
default, matching model); 0/negative values are rejected as invalid input,
consistent with the existing model-alias validation.
