| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The contracts table, supervisor task type, and all their backing
machinery have been inert for several PRs. The directives system reads
its own active contract body for spec text, and PR #135 removed the
last LLM surface that spawned supervisors.
This PR wipes the dead surface in one shot — the user authorised a DB
wipe, so the migration drops every legacy table with CASCADE rather
than carrying forward stub rows. Net change: −12k LOC across handlers,
repository, state, models, the TUI, and the listen module.
What's gone:
- contracts, contract_chat_*, contract_events, contract_repositories,
contract_type_templates tables.
- supervisor_states, supervisor_heartbeats tables.
- mesh_chat_conversations, mesh_chat_messages tables.
- tasks.contract_id/is_supervisor/supervisor_task_id/supervisor_worktree_task_id columns.
- directive_steps.contract_id/contract_type columns.
- files.contract_id/contract_phase columns.
- history_events.contract_id/phase columns.
- The Contract/Supervisor/MeshChat handler + model + repository
surface, plus the daemon TUI views that read them.
- The standalone listen.rs websocket handler (orphaned with the LLM).
What stays:
- mesh_supervisor handler: trimmed to just the questions + orders
backchannel used by `makima directive ask` / `create-order` (kept
the URL prefix for CLI client compat).
- directive_documents (the user-facing "contracts" surface).
- pending_questions in-memory state for the directive Ask flow.
cargo check, cargo test --lib (68 passed), tsc, and vite build all
clean.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
orchestrator (#98)
* feat: soryu-co/soryu - makima: Revert broken directive PRs and verify clean build
* feat: soryu-co/soryu - makima: Re-implement frontend: Lexical document editor with feature flag and base components
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: Add contract blocks, expandable log rows, and interaction controls
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: End-to-end build verification and integration polish
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* fix: resolve compilation error and warnings in Rust backend
- Fix syntax error in directive.rs phase_replanning (bad merge created
duplicate code blocks with broken `.await {` syntax)
- Remove unused imports: WorktreeError, DaemonReauthStatus, ratatui::prelude
- Prefix unused variables with underscore: claude_command, content, owner_id
- Suppress unused_assignments warning on final_exit_code
- Add #[allow(unused_imports)] for cfg(unix) CommandExt imports
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* WIP: heartbeat checkpoint
* fix: suppress remaining compiler warnings for clean build
- Add #[allow(dead_code)] for unused but intentionally kept functions
- Remove useless self-assignments in listen handler
- Fixes: truncate_string, checkout_commit, handle_get_worktree_diff,
default_max_retries, STREAM_CHUNK_MS, listen(), MessageResponse.role
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
| |
* WIP: heartbeat checkpoint
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: Fix daemon reauth flow for new claude setup-token output format
* feat: soryu-co/soryu - makima: Update frontend reconcile toggle to three-way switch
* feat: soryu-co/soryu - makima: Add task slide-out panel to directive page
|
| |
|
|
|
|
|
|
|
|
|
| |
* feat: soryu-co/soryu - makima: Remove aarch64-unknown-linux-gnu from release workflow
* WIP: heartbeat checkpoint
* WIP: heartbeat checkpoint
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: Implement non-blocking ask with client-side polling for reconcile mode
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
improvements (#73)
* feat: soryu-co/soryu - makima: Fix contracts page overflow - constrain layout to viewport height
* feat: soryu-co/soryu - makima: Add git fetch to create_worktree and improve completion prompt merge conflict handling
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: Add pending question notification badge to directive sidebar and nav
* feat: soryu-co/soryu - makima: Fix reconcile:on blocking - make phaseguard poll indefinitely instead of returning immediately
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
bug fixes (#62)
* feat: soryu-co/soryu - makima: Fix directive goal update bug - stale closure issue
* WIP: heartbeat checkpoint
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: Create Orders database schema and backend API
* feat: soryu-co/soryu - makima: Fix task Claude instance not receiving user inputs from input box
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: Build Orders frontend page replacing the Board page
* WIP: heartbeat checkpoint
* WIP: heartbeat checkpoint
* feat: soryu-co/soryu - makima: Fix directive PR creation system
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Restoration
Task 3.3: Supervisor State Persistence
- Add migration 20260201000001_enhanced_supervisor_state.sql with new columns:
- state (supervisor state enum)
- current_activity (description)
- progress (0-100)
- error_message (for failed states)
- spawned_task_ids (tasks created by supervisor)
- pending_questions (questions awaiting user response)
- restoration_count, last_restored_at, restoration_source (restoration tracking)
- Update SupervisorState model with new fields
- Add PendingQuestion struct for tracking unanswered questions
- Add SupervisorRestorationContext for returning restoration info
- Add StateValidationResult and StateRecoveryAction for state validation
State persistence functions in repository.rs:
- update_supervisor_detailed_state() - Update state, activity, progress
- add_supervisor_spawned_task() - Track spawned tasks
- add_supervisor_pending_question() - Track pending questions
- remove_supervisor_pending_question() - Clear answered questions
- save_supervisor_state_full() - Full state save (UPSERT)
- mark_supervisor_restored() - Increment restoration count
- get_supervisors_with_pending_questions() - Find supervisors with pending questions
- get_supervisor_state_for_restoration() - Load state for restoration
- validate_spawned_tasks() - Validate task consistency
- update_supervisor_phase() - Update on phase change
- update_supervisor_heartbeat_state() - Lightweight heartbeat update
State save points:
- On task spawn (save_state_on_task_spawn)
- On question asked (save_state_on_question_asked)
- On question answered (clear_pending_question)
- On phase change (save_state_on_phase_change)
- On heartbeat (update_supervisor_heartbeat_state)
Task 3.4: Supervisor Restoration Protocol
- Add restoration detection when supervisor starts with existing state
- Implement validate_supervisor_state() for state consistency checks
- Implement restore_supervisor() with validation and context generation
- Add redeliver_pending_questions() for re-delivering questions after crash
- Add generate_restoration_context_message() for Claude context injection
- Update resume_supervisor endpoint to return RestorationInfo
- Re-deliver pending questions when supervisor resumes
Restoration flow:
1. Daemon restarts or task reassigned
2. Load supervisor state from supervisor_states
3. If NOT FOUND: Start fresh, log warning
4. If FOUND: Validate state consistency
5. If INVALID: Start from last checkpoint
6. If VALID: Restore conversation history
7. Check for pending questions - re-deliver to user
8. Check for waiting tasks - resume waiting state
9. Send restoration context to Claude
10. Resume execution from last state
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When local_only=true on a contract, all completion actions are skipped.
This adds a new option auto_merge_local that, when enabled along with
local_only, will automatically merge completed task changes to the
master/main branch locally (without pushing or creating PRs).
Changes:
- Add auto_merge_local column to contracts table (migration)
- Add auto_merge_local field to Contract model and summary
- Update CreateContractRequest and UpdateContractRequest structs
- Update contract repository create/update functions
- Add auto_merge_local to WebSocket protocol StartTask command
- Pass auto_merge_local through spawn_task and run_task functions
- Modify task manager completion logic: if local_only=true AND
auto_merge_local=true, execute 'merge' completion action locally
- Update all server handlers to retrieve and pass auto_merge_local
- Add TypeScript types to frontend components
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Task completion checkpoint
* Task completion checkpoint
* Task completion checkpoint
* Fix supervisor merge for completed tasks and make PR command synchronous
## Issue 1: makima supervisor merge doesn't work for completed tasks
When a task completes, the daemon removes it from in-memory task tracking.
This caused merge operations to fail with "Task not found".
Fixed by updating handle_merge_task_to_target() to use get_task_worktree_path()
which scans the worktrees directory as a fallback when the task is not in memory.
Also updated handle_create_pr() with the same pattern for consistency.
## Issue 2: makima supervisor pr returns immediately without result
The create_pr handler was asynchronous - it sent the CreatePR command to the
daemon and immediately returned without waiting for the result.
Fixed by:
1. Adding PrResultNotification struct and pr_results broadcast channel to AppState
2. Updating mesh_daemon.rs to broadcast PRCreated results to the channel
3. Updating create_pr() handler to subscribe to pr_results and wait for the
result with a 60-second timeout (matching the merge command pattern)
Now the PR command returns the actual pr_url and pr_number from the daemon.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Implements a parallel "red team" task that monitors work task outputs in
real-time, verifying implementations stick to contract requirements,
repository standards, and the execution plan.
Key features:
- New `red_team_enabled` and `red_team_prompt` contract configuration
- Red team tasks auto-spawn when first work task is created
- `makima red-team notify` CLI command for alerting supervisors
- POST /api/v1/mesh/red-team/notify and /status endpoints
- Alert delivery to supervisor via SendMessage daemon command
- Notification audit trail via history_events table
Database changes:
- Add red_team_enabled/red_team_prompt columns to contracts
- Add is_red_team flag to tasks with partial index
- Create red_team_notifications table for audit logging
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add local_only flag to contracts database and models
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* Task completion checkpoint
* Skip automatic completion actions in local_only mode
Add `local_only` flag to contracts that prevents automatic completion
actions (branch, merge, pr) from executing when tasks complete. This
allows users to manually handle code changes via patch files or other
means when operating in local-only mode.
Changes:
- Add `local_only` field to Contract model and request types
- Add database migration for the new column
- Add `local_only` parameter to SpawnTask command in both state.rs and
daemon protocol.rs
- Modify task manager to skip completion action execution when
`local_only` is true, with appropriate logging
- Pass `local_only` flag through all task spawning paths:
- mesh_supervisor.rs (task spawn, retry, resume)
- mesh.rs (task start, reassign, continue)
- mesh_chat.rs (run task)
- contract_chat.rs (run task)
- Update repository create/update functions to handle `local_only`
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* Task completion checkpoint
* Implement core patch export system
Add functionality to create uncompressed, human-readable git patches
for export. This enables users to generate patches that can be manually
applied or shared, without the compression used for internal checkpoints.
Changes:
- Add ExportPatchResult struct with patch content, file count, and line stats
- Add create_export_patch() function that generates diffs against a base SHA
- Add get_head_sha() utility function
- Add parse_diff_stat() helper to extract line counts from git output
- Add CreateExportPatch command to daemon protocol
- Add ExportPatchCreated response message to protocol
- Add handler in task manager to process export patch requests
- Add server-side handling to broadcast patch results to UI
The export patch system automatically finds the merge-base when no base
SHA is provided, trying upstream tracking branch first, then common
default branches (origin/main, origin/master, main, master).
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* Task completion checkpoint
* Add GitActionsPanel frontend component
* Add WorktreeFilesPanel and PatchesListPanel components
* Add local-only mode toggle to contract creation
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
* [WIP] Heartbeat checkpoint - 2026-01-20 22:40:37 UTC
* Task completion checkpoint
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add phaseguard flag to supervisor ask command
When phaseguard is enabled (--phaseguard flag), the supervisor's ask
command will block indefinitely waiting for user response instead of
timing out after the default 1 hour. This ensures that phase transitions
requiring user confirmation cannot proceed without explicit approval.
Changes:
- Add phaseguard field to AskQuestionRequest in server handler
- Add phaseguard field to AskQuestionRequest in API client
- Add --phaseguard CLI flag to AskArgs
- Update supervisor_ask() to accept phaseguard parameter
- Modify ask_question handler to skip timeout check when phaseguard=true
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* Task completion checkpoint
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
|
| | |
|
| | |
|
| |
|
|
| |
available
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This PR implements a comprehensive resume and history system that enables:
1. **History Viewing**
- View complete conversation history for contracts across all phases
- View conversation history for individual tasks
- View task output/tool call history with timestamps
- View checkpoint history
- Timeline view showing all activities
2. **Resume System**
- Resume interrupted supervisor conversations with full context
- Resume interrupted task conversations
- Resume from specific checkpoints
- Continue tasks from previous task state (worktree inheritance)
3. **Rewind/Restore Features**
- Rewind code to any checkpoint (git restore)
- Rewind conversation to any point
- Create new branches from historical points
- Fork tasks from any point in history
- New migration: 20250117000000_history_tables.sql
- conversation_snapshots table for storing conversation state
- history_events table for unified timeline
- Added forking fields to tasks table
- Added conversation_snapshot_id to task_checkpoints
- ConversationSnapshot, HistoryEvent, ConversationMessage
- Request/response types for resume and rewind operations
- Query filter types for history endpoints
- CRUD functions for conversation_snapshots
- CRUD functions for history_events
- Task conversation retrieval from task_events
- GET /api/v1/contracts/{id}/history
- GET /api/v1/contracts/{id}/supervisor/conversation
- GET /api/v1/mesh/tasks/{id}/conversation
- GET /api/v1/timeline
- POST /api/v1/contracts/{id}/supervisor/resume
- POST /api/v1/mesh/tasks/{id}/rewind
- POST /api/v1/mesh/tasks/{id}/fork
- POST /api/v1/mesh/tasks/{id}/checkpoints/{cid}/resume
- POST /api/v1/mesh/tasks/{id}/checkpoints/{cid}/branch
- POST /api/v1/contracts/{id}/supervisor/conversation/rewind
- task-history: View task conversation history
- task-checkpoints: List task checkpoints
- resume: Resume supervisor after interruption
- task-resume-from: Resume task from checkpoint
- task-rewind: Rewind task code to checkpoint
- task-fork: Fork task from historical point
- rewind-conversation: Rewind supervisor conversation
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add comprehensive CLI documentation
- Create makima/docs/CLI.md with complete command reference for:
- makima server: HTTP/WebSocket server options
- makima daemon: Worker daemon configuration
- makima supervisor: Contract orchestration commands
- makima contract: Task-contract interaction commands
- Include configuration file examples and environment variables
- Add usage workflows for common scenarios
- Update makima/README.md with CLI overview and link to docs
Add GitHub Actions release workflow for v0.1.0
Creates automated release workflow that:
- Triggers on v* tag pushes
- Builds binaries for Linux x86_64, macOS x86_64, and macOS ARM64
- Uses Rust nightly toolchain (required for edition 2024)
- Packages binaries as .tar.gz archives
- Creates GitHub release with installation instructions
fix(ci): update macOS runner for x86_64 builds
Replace deprecated macos-13 runner with macos-15-intel for
x86_64-apple-darwin target. The macos-13 runner has been retired
by GitHub Actions.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add dismissing notifications and fix CLI task ID arg
Add worktree cleanup when contracts complete or are deleted (#21)
- Add CleanupWorktree daemon command variant
- Handle CleanupWorktree in daemon task manager
- Add cleanup_contract_worktrees helper function
- Trigger cleanup when contract status becomes 'completed'
- Trigger cleanup before contract deletion
Add Autonomous Loop Mode for persistent task completion (#20)
Implements the "Autonomous Loop Mode" feature inspired by Ralph for Claude Code.
This enables tasks to automatically restart and continue working until they
explicitly signal completion via a COMPLETION_GATE block.
Key features:
- Exit confirmation via COMPLETION_GATE: Tasks must output a <COMPLETION_GATE>
block with `ready: true` to signal completion. Without this, the task
auto-restarts using `claude --continue` to resume the conversation.
- Circuit breaker: Prevents infinite loops by detecting:
* Maximum iteration limit (default: 10)
* No progress for N consecutive iterations (default: 3)
* Same error repeated N times (default: 5)
- spawn_continue: New ProcessManager method to spawn Claude with the
`--continue` flag, resuming from the previous session state.
Toggle: Enable via `autonomous_loop` flag on contracts. When set, all tasks
spawned for that contract will run in autonomous loop mode.
Files changed:
- completion_gate.rs: COMPLETION_GATE parser and CircuitBreaker logic
- claude.rs: spawn_continue() for --continue mode spawning
- manager.rs: Autonomous loop iteration logic in run_task()
- protocol.rs: autonomousLoop field in DaemonCommand::SpawnTask
- models.rs/repository.rs: autonomous_loop column on contracts/tasks
- Migration: Adds autonomous_loop columns to contracts and tasks tables
Add get-task and output commands to supervisor CLI (#24)
Add two new supervisor subcommands:
- `makima supervisor task <task_id>` - Get individual task details
- `makima supervisor output <task_id>` - Get task output/claude log
This allows supervisors to fetch task details and claude output
directly from the CLI instead of using curl to call the task API.
Add optional bubblewrap sandboxing for Claude processes (#23)
Add --bubblewrap flag and process.bubblewrap config section to enable
running Claude Code in a bubblewrap sandbox for process isolation.
When enabled, claude processes run with filesystem restrictions:
- Root filesystem mounted read-only
- Working directory (worktree) mounted read-write
- Fresh /dev, /proc, /tmp
- Network access preserved for API calls
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
|
| | |
|
| |
|
