| Age | Commit message (Collapse) | Author |
|---|
|
UpdateTaskState now validates the transition using ValidTransition inside
a transaction. Invalid transitions return an error rather than blindly
updating. Tests for retry-limit and running-task-rejection test setup
are updated to create tasks with the target state directly via CreateTask
to bypass the transition guard in setup code.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
- ClaudeConfig.WorkingDir → ProjectDir (json: project_dir)
- UnmarshalJSON fallback reads legacy working_dir from DB records
- New executions with project_dir clone into a temp sandbox via git clone --local
- Non-git project_dirs get git init + initial commit before clone
- After success: verify clean working tree, merge --ff-only back to project_dir, remove sandbox
- On failure/BLOCKED: sandbox preserved, path included in error message
- Resume executions run directly in project_dir (no re-clone)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Add --dirty flag to skip stashing behavior. Stashing includes untracked
files and uses a trap to ensure changes are popped back on exit.
|
|
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
- Running tab in nav with live SSE log streams per running task
- Execution history table (last 24h) with duration, cost, exit code, view logs
- Poll loop refreshes running view when tab is active
- Smart diff: only full re-render when task set changes; elapsed updated in place
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
When claude hits the 5-hour usage limit it exits 1. execOnce was
returning the generic "exit status 1" error, hiding the real cause from
the retry loop and the task state machine.
Fix:
- execOnce now surfaces streamErr when it indicates rate limiting or
quota exhaustion, so callers see the actual message.
- New isQuotaExhausted() detects "hit your limit" messages — these are
not retried (retrying a depleted 5h bucket wastes nothing but is
pointless), and map to BUDGET_EXCEEDED in both execute/executeResume.
- isRateLimitError() remains for transient throttling (429/overloaded),
which continues to trigger exponential backoff retries.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
subtasks
Replaces the API POST instructions with the claudomator create command,
which is simpler and consistent with how operators queue tasks.
--start is explicitly omitted so subtasks are queued but not auto-started.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
claudomator create <name> -i <instructions> [flags]
--parent-id attach as subtask of given task ID
--working-dir working directory
--model claude model
--budget max USD
--timeout task timeout
--priority high/normal/low
--start queue immediately after creating
deploy script now also copies binary to /usr/local/bin/claudomator.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
The elaborate call now sends working_dir from the Project dropdown.
The backend uses it (falling back to server workDir) when building
the system prompt, so AI-drafted tasks are contextualised to the
selected project.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
/workspace/claudomator
- Moved working directory to first field, renamed to "Project"
- Replaced text input with a select populated from GET /api/workspaces
(lists subdirs of /workspace dynamically)
- "Create new project…" option reveals a custom path input
- elaborate result handler sets select or falls back to new-project input
- Added GET /api/workspaces endpoint in server.go
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Recovered files with no Claude→Agent contamination:
- docs/adr/002-task-state-machine.md
- internal/api/logs.go/logs_test.go: task-level log streaming endpoint
- internal/api/validate.go/validate_test.go: POST /api/tasks/validate
- internal/api/server_test.go, storage/db_test.go: expanded test coverage
- scripts/reset-failed-tasks, reset-running-tasks
- web/app.js, index.html, style.css: frontend improvements
- web/test/: active-tasks-tab, delete-button, filter-tabs, sort-tasks tests
Manually applied from server.go diff (skipping Claude→Agent rename):
- taskLogStore field + validateCmdPath field
- DELETE /api/tasks/{id} route + handleDeleteTask
- GET /api/tasks/{id}/logs/stream route
- POST /api/tasks/{id}/resume route + handleResumeTimedOutTask
- handleCancelTask: allow cancelling PENDING/QUEUED tasks directly
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
restart
Two bugs:
1. SubmitResume was called with r.Context(), which is cancelled as soon
as the HTTP handler returns, immediately cancelling the resume execution.
Switch to context.Background() so the execution runs to completion.
2. CANCELLED→QUEUED was missing from ValidTransition, so the Restart
button on cancelled tasks always returned 409. Added the transition.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
session_id was set on the Execution struct inside ClaudeRunner.Run but
was missing from the UPDATE query, so it was never written to the DB.
GetLatestExecution then returned an empty session_id, causing the
/answer endpoint to return "no resumable session found".
Also patched the existing blocked execution row directly in the DB.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Tapping the answer input or question text on mobile bubbled up to the
card's click handler, opening the detail panel and stealing focus.
Stopping propagation at the footer level covers all child elements.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
READY state means a parent task whose subtasks should now run — select
its first PENDING child, not a global fallback. COMPLETED state means
continue the series via the next PENDING sibling. Only fall back to the
priority+age queue when neither yields a task.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
When the most recently completed task had no remaining PENDING siblings
(series fully done), the script returned empty and start-next-task
reported "No task to start." Fix by falling through to the
priority+age fallback whenever the sibling search yields nothing.
Also add header comment documenting the selection logic, and suppress
browser favicon 404s with <link rel="icon" href="data:,">.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
POST /api/tasks/{id}/cancel now works. Pool tracks a cancel func per
running task ID; Cancel(taskID) calls it and returns false if the task
isn't running. The execute goroutine registers/deregisters the cancel
func around the runner call.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
When an agent needs user input it writes a question to
$CLAUDOMATOR_QUESTION_FILE and exits. The runner detects the file and
returns BlockedError; the pool transitions the task to BLOCKED and
stores the question JSON on the task record.
The user answers via POST /api/tasks/{id}/answer. The server looks up
the claude session_id from the most recent execution and submits a
resume execution (claude --resume <session-id> "<answer>"), freeing the
executor slot entirely while waiting.
Changes:
- task: add StateBlocked, transitions RUNNING→BLOCKED, BLOCKED→QUEUED
- storage: add session_id to executions, question_json to tasks;
add GetLatestExecution and UpdateTaskQuestion methods
- executor: BlockedError type; ClaudeRunner pre-assigns --session-id,
sets CLAUDOMATOR_QUESTION_FILE env var, detects question file on exit;
buildArgs handles --resume mode; Pool.SubmitResume for resume path
- api: handleAnswerQuestion rewritten to create resume execution
- preamble: add question protocol instructions for agents
- web: BLOCKED state badge (indigo), question text + option buttons or
free-text input with Submit on the task card footer
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
claudomator runs tasks unattended; prompting for write permission
always stalls execution. Any task without an explicit permission_mode
now gets --permission-mode bypassPermissions automatically.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Add LogPather interface; ClaudeRunner implements it via ExecLogDir().
Pool pre-populates stdout_path/stderr_path/artifact_dir on the execution
record before CreateExecution, so paths are in the DB from the moment
a task starts running.
ClaudeRunner.Run() skips path assignment when already set by the pool.
Also update scripts/debug-execution to derive paths from the known
convention (<data-dir>/executions/<exec-id>/) as a fallback for
historical records that predate this change.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
restartTask() was POSTing to /api/tasks/{id}/restart (405) instead
of the existing /api/tasks/{id}/run endpoint.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Rename streamAndParseCost → parseStream (returns float64, error).
Detect two failure modes that claude reports via exit 0:
- result message with is_error:true
- tool_result permission denial ("haven't granted it yet")
Also fix cost extraction to read total_cost_usd from the result
message (the actual field name), keeping cost_usd as legacy fallback.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
- READY state task cards show Accept + Reject buttons
- Accept POSTs to /api/tasks/{id}/accept (→ COMPLETED)
- Reject POSTs to /api/tasks/{id}/reject (→ PENDING)
- "Start Next" button in toolbar POSTs to /api/scripts/start-next-task
- CSS for .btn-accept and .btn-reject
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Explicitly state that /workspace/claudomator is the canonical directory.
Prevents agents from wandering into other directories on the filesystem.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Merges features developed in /site/doot.terst.org/claudomator-work (a
stale clone) into the canonical repo:
- executor: QuestionRegistry for human-in-the-loop answers, rate limit
detection and exponential backoff retry (ratelimit.go, question.go)
- executor/claude.go: process group isolation (SIGKILL orphans on cancel),
os.Pipe for reliable stdout drain, backoff retry on rate limits
- api/scripts.go: POST /api/scripts/start-next-task handler
- api/server.go: startNextTaskScript field, answer-question route,
BroadcastQuestion for WebSocket question events
- web: Cancel/Restart buttons, question banner UI, log viewer, validate
section, WebSocket auto-connect
All tests pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
|
|
debug-execution: inspect a failed execution by ID prefix from prod DB.
deploy: build and restart the claudomator systemd service.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Adds `claudomator start <task-id>` to queue a task via the running API
server. Adds internal/version for embedding VCS commit hash in the
server banner.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
next-task now outputs only the task ID (or nothing), removing prose
prefixes that made downstream parsing fragile. start-next-task
simplifies to a direct empty-check with no awk required.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Top-level tasks now land in READY after successful execution instead of
going directly to COMPLETED. Subtasks (with parent_task_id) skip the gate
and remain COMPLETED. Users accept or reject via new API endpoints:
POST /api/tasks/{id}/accept → READY → COMPLETED
POST /api/tasks/{id}/reject → READY → PENDING (with rejection_comment)
- task: add StateReady, RejectionComment field, update ValidTransition
- storage: migrate rejection_comment column, add RejectTask method
- executor: route top-level vs subtask to READY vs COMPLETED
- api: /accept and /reject handlers with 409 on invalid state
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
executor/claude.go: stat working_dir before cmd.Start() so a missing
or inaccessible directory surfaces as a clear error
("working_dir \"/bad/path\": no such file or directory") rather than
an opaque chdir failure wrapped in "starting claude".
api/elaborate.go: replace the hardcoded /root/workspace/claudomator
path with buildElaboratePrompt(workDir) which injects the server's
actual working directory (from os.Getwd() at startup). Empty workDir
tells the model to leave working_dir blank.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
- Tab bar to switch between Tasks and Templates views
- New Task modal with elaborate section ("Draft with AI") that calls
POST /api/tasks/elaborate and pre-fills form fields
- Templates panel listing saved configs with create/edit/delete
- base-path meta tag for sub-path deployments
- filter.test.mjs: contract tests for filterTasks function
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
- POST /api/tasks/elaborate: calls claude to draft a task config from
a natural-language prompt
- GET /api/executions/{id}/logs/stream: SSE tail of stdout.log
- CRUD /api/templates: create/list/get/update/delete reusable task configs
- GET /api/tasks/{id}/subtasks: list child tasks
- Server.NewServer accepts claudeBinPath for elaborate; injectable
elaborateCmdPath and logStore for test isolation
- Valid-transition guard added to POST /api/tasks/{id}/run
- CLI passes claude binary path through to the server
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
- Pool.waitForDependencies polls depends_on task states before running
- ClaudeRunner prepends planningPreamble to task instructions to prompt
a plan-then-implement approach
- Rate-limit test helper updated to match new ClaudeRunner signature
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
- Task struct gains ParentTaskID field
- DB schema adds parent_task_id column (additive migration)
- DB.ListSubtasks fetches children of a parent task
- DB.UpdateTask allows partial field updates (name, description, state, etc.)
- Templates table added to initial schema
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Add patterns for compiled test binaries (*_test_bin, *.test), test
output files, nohup.out, overlay.json, the tasks/ data dir, and the
scratch binary x.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Replace the static "N hidden tasks" label with a toggle button that
expands an inline fold showing the hidden task cards dimmed at 0.6
opacity. Fold state is module-level so it survives poll cycles within
a session but resets on reload.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Documents build/test commands, package roles, key data flows, task
YAML format, storage schema, and ADR references for AI-assisted
development sessions.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Serves a lightweight dashboard (HTML/CSS/JS) from an embedded FS at
GET /. The JS polls the REST API to display tasks and stream logs
via WebSocket. Static files are embedded at build time via web/embed.go.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Adds `claudomator logs <execution-id>` to stream or display stdout logs
from a past or running execution. Includes unit tests for the command.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Ensures richer stream-json output for cost parsing and debugging.
Adds a test to verify --verbose is always present in built args.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Without json tags, Go serialized fields with capitalized names (State,
CreatedAt, Name) but the UI expected snake_case (task.state,
task.created_at). This caused createTaskCard to throw TypeError when
tasks existed, which poll() caught and displayed as "Could not reach
server".
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
|
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
|
Claudomator automation toolkit for Claude Code with:
- Task model with YAML parsing, validation, state machine (49 tests, 0 races)
- SQLite storage for tasks and executions
- Executor pool with bounded concurrency, timeout, cancellation
- REST API + WebSocket for mobile PWA integration
- Webhook/multi-notifier system
- CLI: init, run, serve, list, status commands
- Console, JSON, HTML reporters with cost tracking
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
