* feat(agentctl): implement protocol adapter abstraction
- Add AgentAdapter interface for protocol-agnostic agent communication
- Create ACPAdapter implementation for ACP protocol (JSON-RPC over stdin/stdout)
- Add Protocol field to agent registry (AgentTypeConfig)
- Update process.Manager to use adapters instead of hardcoded ACP
- Update API layer to use adapter interface
- Pass protocol from registry to agentctl via CLI flag and env var
- Support protocol configuration in both Docker and standalone modes
This refactoring allows agentctl to support multiple protocols (ACP, REST, MCP)
via a pluggable adapter pattern. The backend remains protocol-agnostic,
always communicating with agentctl's HTTP API.
* refactor(agentctl): consolidate types and clean up code
- Create shared types package (internal/agentctl/types) for permission types
- Create shared protocol package (pkg/agent) for Protocol type
- Remove duplicate permission types from acp/client.go and adapter.go
- Make adapter the single source of truth for session ID
- Remove unused Command field from adapter.Config
- Remove passthrough wrapper function handleAdapterPermissionRequest
- Clean up extra empty lines in process/manager.go and acp_adapter.go
- Re-export permission types from adapter package for convenience
* fix(agent-session): rename AgentType to AgentProfileID for consistency
- Rename AgentSession.AgentType to AgentProfileID in models
- Update agent_sessions table schema column from agent_type to agent_profile_id
- Update all SQL queries in sqlite.go to use agent_profile_id
- Fix ToAPI() method to return agent_profile_id key
- Remove misleading comments in executor.go
Also adds comprehensive tests for affected areas:
- adapter package unit tests (SessionUpdate, PlanEntry, AgentInfo, Config)
- profile resolver unit tests with mock repository
- AgentSession CRUD tests in sqlite_test.go
* feat: replace agent_type with agent_profile_id
- Replace agent_type field with agent_profile_id across backend and frontend
- Add ProfileResolver interface and StoreProfileResolver implementation
- Update task creation/update to use agent_profile_id
- Change default agent runtime from docker to standalone
- Fix workspace state types to include all required fields
- Fix websocket handlers type mismatches for environments, executors, workspaces
- Update integration tests to use agent_profile_id
* feat: pass model as CLI argument to agents
- Add ModelFlag field to AgentTypeConfig for specifying the CLI flag
- Set ModelFlag to '--model' for all agents (auggie, claude, codex, gemini, opencode, copilot)
- Update launchStandalone to append model flag to agent command
- Update launchDocker/buildContainerConfig to append model flag to container command
- Model is passed when both agent supports ModelFlag and profile has Model configured
* fix: make WebSocket message handling async to prevent blocking
When a long-running handler like orchestrator.prompt was executing,
it blocked the ReadPump loop, preventing other requests (like
workspace.tree.get) from being processed until the handler completed.
Now each message is handled in its own goroutine, allowing concurrent
message handling so file tree requests work while the agent is working.
* feat: show task description as first message in chat
- Add 'Task' label with distinct styling (lighter background, border)
- Pass taskDescription from page to TaskChatPanel component
- Differentiate from regular user messages visually
* feat(web): add file browser with syntax-highlighted viewer
- Add file browser component in right panel to browse workspace files
- Open files in left panel tabs alongside Notes/Changes/Chat
- Color-coded dots for file types (TS, JS, Go, Python, etc.)
- Visual separator between main tabs and file tabs
- Folders sorted to top in file tree view
- LRU eviction for max 4 open file tabs
- Backend handlers for workspace file tree and content via WebSocket
* fix(web): refresh file tree on filesystem changes via fsnotify
- Subscribe to file change notifications from agentctl's fsnotify watcher
- Backend lifecycle manager now connects to file changes stream
- Publishes FileChangeNotified events to event bus
- WebSocket broadcasts workspace.file.changes to task subscribers
- Frontend file browser refreshes tree when changes detected
* perf(agentctl): use fsnotify instead of polling for git status
- Replace 2-second polling with fsnotify-triggered updates
- Add debouncing (300ms) to batch rapid file changes
- Git status updates now triggered by actual filesystem changes
- More efficient and responsive than periodic polling
* refactor(agentctl): consolidate fs events into single debounced trigger
- Single fsChangeTrigger channel for all filesystem events
- After 300ms debounce: update git status, files, and notify subscribers
- Removed per-file notifications in favor of single 'refresh' event
- Simpler and more efficient than separate triggers
* refactor(agentctl): simplify file API - use HTTP for tree/content
- Remove unused filesConn, fileTreeConn, fileContentConn from Client struct
- Convert RequestFileTree and RequestFileContent from WebSocket to HTTP GET
- Remove StreamFiles and CloseFilesStream (redundant with file changes stream)
- Keep only fileChangesConn for the debounced refresh notification
HTTP is simpler than WebSocket for request/response patterns.
* feat(web): switch to CodeMirror for file viewer
- Replace react-syntax-highlighter with @uiw/react-codemirror
- Add language extensions for JS/TS, Python, Go, Rust, Java, C++, CSS, HTML, JSON, Markdown, YAML
- Use VS Code dark theme for consistent look
- Enable code folding and read-only mode
- Better performance and features than syntax highlighter
* refactor: consolidate duplicated types and extract shared constants
- Re-export file types from process package in agentctl/client.go instead of duplicating
- Move OpenFileTab type to lib/types/backend.ts (shared by file-browser and page-client)
- Extract FILE_EXTENSION_COLORS constant for file type color indicators
- Remove console.log statement and unused imports in file-browser.tsx
- Add 'refresh' operation to FileChangeNotificationPayload type
* refactor(agentctl): move Close() method from acp.go to client.go
Close() is a general client cleanup method that closes all streams
(ACP, permissions, git status, file changes). It belongs in client.go
since acp.go should only contain ACP-specific communication logic.
* fix(web): remove unused cn import in file-browser
* refactor(agentctl): consolidate duplicated types into process/types.go
- Create process/types.go with all workspace-related types
- Remove duplicate GitStatusUpdate, FileInfo from workspace_tracker.go
- Remove duplicate GitStatusUpdate, FileInfo from api/workspace.go
- Re-export all types from agent/agentctl/client.go for convenience
Types are now defined once and imported everywhere.
* chore: remove package-lock.json, add to .gitignore
Project uses pnpm (pnpm-lock.yaml), not npm.
* fix(orchestrator): probe agent before returning stale execution state
When checking task.execution, the backend now probes the actual agent:
- For tracked instances: health check via agentctl
- For Docker: verify container is running
If the agent is not running but DB says 'running', the session is
marked as stopped and false is returned. This fixes the UI showing
agents as running after backend restart when they're actually dead.
Changes:
- Add IsAgentRunningForTask to AgentManagerClient interface
- Add IsContainerRunning to docker client
- Update GetExecution to verify agent is actually running
- Clean up stale sessions in database when agent not found
* fix(orchestrator): use agentctl status to verify agent is running
Simplified approach - probe agentctl's /api/v1/status endpoint to check
if agent process is running. Same check for both Docker and standalone.
- Add IsAgentRunning() helper to StatusResponse
- Simplify IsAgentRunningForTask to just use agentctl.GetStatus()
* fix(orchestrator): also verify agent running for in-memory executions
The previous fix only checked agent status when loading from DB. But if
an execution is cached in memory and the agent process stops (while
agentctl stays running), the stale state was still returned.
Now both paths verify the agent is actually running by probing agentctl.
If agent stopped, clean up in-memory state and update DB session.
* chore(web): remove unused react-syntax-highlighter
Switched to CodeMirror for code viewing, no longer needed.
* fix(tests): add IsAgentRunningForTask to mock agent managers
Add the new interface method to test mocks:
- mockAgentManager in scheduler_test.go
- SimulatedAgentManagerClient in orchestrator_test.go
---------
Co-authored-by: Kandev Agent <kandev@example.com>
* feat: implement real-time git status integration
- Add workspace tracker to monitor git changes every 2 seconds
- Implement WebSocket streaming for git status updates
- Add database persistence for git status in agent session metadata
- Compute diffs backend-side with additions/deletions per file
- Integrate git status into frontend UI (top bar, files panel, changes panel)
- Display real-time ahead/behind commits and line changes
- Show file diffs with syntax highlighting in task view
- Add retry logic for WebSocket connections during reconciliation
- Fix context lifecycle issue in workspace tracker
* refactor: remove debug and unnecessary logging
- Remove debug logs from workspace tracker
- Remove info logs from WebSocket handlers
- Remove verbose git status update logs
- Keep only essential error logs
* fix: compare git diff against remote branch instead of HEAD
- Changed git diff comparison from HEAD to remote tracking branch
- Now shows all changes including committed files that haven't been pushed
- Files committed locally but not pushed will appear in the diff
- Falls back to HEAD if no remote branch is configured
* fix: prevent stale git status when switching tasks
- Clear git status when switching to a new task
- Only accept git.status WebSocket updates for the current task
- Prevents showing git data from previous task when switching
* refactor: remove unused diff stream client code
- Remove diffConn field from agentctl Client struct
- Remove StreamDiff() method (never called)
- Remove CloseDiffStream() method
- Remove DiffUpdate type from client.go
- All diff data is already included in GitStatusUpdate.Files
- Keep server-side diff endpoint in agentctl for potential future use
* refactor: remove unused diff stream server-side code
- Remove /workspace/diff/stream endpoint from agentctl API
- Remove handleDiffStreamWS handler
- Remove DiffUpdate type from workspace_tracker.go
- Remove DiffSubscriber type
- Remove updateDiff() and getDiff() methods
- Remove SubscribeDiff(), UnsubscribeDiff() methods
- Remove notifyDiffSubscribers() method
- Remove GetCurrentDiff() method
- Remove currentDiff field and diffSubscribers map
- All diff data is already included in GitStatusUpdate.Files
* refactor: remove unused output stream code
- Remove /output and /output/stream endpoints from agentctl API
- Remove handleGetOutput and handleOutputStreamWS handlers
- Delete output.go file entirely
- Remove OutputBuffer and related code (buffer.go deleted)
- Remove outputConn field from agentctl Client
- Remove StreamOutput() and CloseOutputStream() methods
- Remove OutputLine type from client.go
- Remove OutputBufferSize config option
- Update readStderr to just log instead of buffering
Output data flows through ACP stream (StreamUpdates) which includes
all session notifications. The separate output stream was never used.
- Add reconnectStreams() to reconnect WebSocket streams for recovered agents
- Reconnect updates stream (for agent responses)
- Reconnect permissions stream (for permission requests)
- Mark recovered agents as READY so they can accept prompts
- Fix worktree cleanup order: clean files before DB cascade deletes records
- Add WorktreePath and WorktreeBranch to StartTaskResponse
- Add those fields to TaskExecution and LaunchAgentResponse
- Store worktree_path and worktree_branch in AgentInstance metadata
- Update frontend to use worktree info from start response directly
Container runs as root but mounted files are owned by host user.
Git 2.35.2+ rejects this as 'dubious ownership'. Configure git via
environment variables to trust all directories in the container.
Git worktree metadata in .git/worktrees/{name} has a 'commondir' file
that references back to the parent .git directory (typically '../..').
Mounting just the worktrees subdirectory wasn't sufficient - we need
to mount the entire .git directory so that:
1. The worktree's .git file can resolve to .git/worktrees/{name}
2. The commondir reference can resolve back to the main .git
3. Git objects, refs, and other shared data are accessible
Git worktrees store their metadata in the parent repository's
.git/worktrees/{name} directory. The worktree's .git file contains
an absolute path to this directory.
When mounting a worktree into a container, git commands fail because
the .git file references a host path that doesn't exist in the container.
This fix mounts the git worktree metadata directory into the container
at the same path it exists on the host, so git commands work correctly.
- Add worktree package with Manager, Store, Config, and error types
- Integrate worktree creation/reuse into lifecycle manager
- Update executor to pass worktree configuration
- Add WorktreeConfig to common config package
- Enable worktrees by default for agent isolation
- Add unit tests for worktree manager
This allows multiple agents to work on the same repository simultaneously
by creating isolated worktrees for each task. Worktrees persist between
agent runs for task resumption and are only cleaned up when tasks are
explicitly deleted.
- Add typing indicator in TaskChatPanel that shows when agent is working
- Enhanced ToolCallCard component with expandable details section
- Backend now passes tool call args (kind, path, locations, raw_input) to comments
- Add generateUUID fallback for non-secure contexts (HTTP non-localhost)
- Configure Next.js allowedDevOrigins for cross-origin dev access
- Bind dev server to 0.0.0.0 for network access
Backend changes:
- Extract Kind, Locations, RawInput from ACP ToolCall updates
- Add Args field to ToolCallData in watcher
- Pass args through CommentCreator interface to comment metadata
Frontend changes:
- ToolCallCard shows file path, status icons, expandable args
- getToolIcon matches ACP ToolKind values (read, edit, search, execute)
- Typing indicator with animated dots when awaiting agent response
- Replace crypto.randomUUID with fallback utility across all files
- Add Docker reconciliation on lifecycle manager startup to find running containers
- Add container_id field to agent_sessions table for cleanup tracking
- Add GetRecoveredInstances to AgentManagerClient interface
- Sync executor cache with recovered Docker instances on startup
- Add task.execution WebSocket endpoint to check agent running state
- Fix Stop Agent button showing Start Agent after page refresh
- Ensure containers are not leaked when backend restarts
- Remove validation requiring repository_url in lifecycleAdapter.LaunchAgent
- Skip workspace mounts in lifecycle manager when no workspace path is provided
- Agents can now run without a pre-existing workspace directory
- Add AgentSession database entity with SQLite persistence
- Implement AgentSession CRUD operations in repository layer
- Update executor to use hybrid cache + database approach
- Add session restoration on startup (LoadActiveSessionsFromDB)
- Link comments to agent sessions via agent_session_id
- Fix SQLite NULL handling for nullable task fields
- Add dev database setup with seed data (make dev, make dev-db-reset)
- Update Makefile with dev targets for development mode
This enables agent sessions to persist across backend restarts,
solving the 'execution not found' error for resumed conversations.
- Add comment handlers for WebSocket API (comment.add, comment.list)
- Update e2e-test.sh to focus on comment system testing
- Test multi-turn conversation flow (3 turns)
- Display comment content and metadata in test output
- Reduce test timeouts for faster execution
- Fix various orchestrator and lifecycle manager issues
- Add CODE_REVIEW_ISSUES.md documenting known issues
- Add permission stream support to backend agentctl client
- StreamPermissions() connects to permission WebSocket
- RespondToPermission() sends responses via HTTP POST
- PermissionNotification type for permission request data
- Connect lifecycle manager to permission stream
- handlePermissionStream() receives permission requests
- publishPermissionRequest() forwards to event bus
- RespondToPermission() and RespondToPermissionByTaskID() methods
- Add WebSocket handler for permission.respond action
- Validates task_id, pending_id, option_id
- Routes response through orchestrator → executor → lifecycle manager
- Add permission WebSocket actions
- ActionPermissionRequested for notifications
- ActionPermissionRespond for user responses
- Update executor interface with RespondToPermissionByTaskID()
- Update mock implementations in tests
- Add permission response test to e2e-test.sh
- Fix e2e-test.sh ID extraction (strip newlines)
The full permission flow is now wired up:
agent → agentctl → backend → frontend → user → backend → agentctl → agent
Auto-approve mode is still enabled by default until frontend is ready.
- Add Comment model and CRUD operations (repository, service layers)
- Add WAITING_FOR_INPUT task state for agent input requests
- Add WebSocket actions for comment.add, comment.get, comment.list
- Add ActionCommentAdded and ActionInputRequested notifications
- Update PromptAgent to return PromptResult with stop reason
- Add e2e test for agent input request flow
- Add PR template for the repository
This enables agents to request input from users by returning
stopReason 'needs_input' which transitions the task to
WAITING_FOR_INPUT state and creates an agent comment.
The session_id must be in the 'data' field, not at the top level,
so the watcher can parse it into protocol.Message and the handler
can find it at msg.Data["session_id"].
The lifecycle manager now publishes a session_info event after creating
an ACP session. This event is handled in main.go to store the session ID
in the task metadata, enabling session resumption for follow-up tasks.
- Add publishSessionNotification() to lifecycle manager
- Convert agentctl SessionNotification to event bus message
- Enables real-time streaming of agent output to WebSocket clients
- Add AgentStatusReady status for when agent finishes a prompt
- Add AgentReady event published when prompt completes
- PromptAgent accepts RUNNING/READY states, marks READY when done
- Add /complete endpoint to explicitly complete a task
- E2E test polls for READY status instead of file creation
- E2E test validates multi-turn conversation flow
- Use acp-go-sdk for all ACP communication in agentctl
- Mark agent as completed when Prompt() returns successfully
- Update task state to COMPLETED when agent completes
- Update task state to FAILED when agent fails and retries exhausted
- Enhanced E2E test to verify agent actually creates files
- E2E test now validates full flow: task creation -> agent work -> file creation -> task completion
Carlos Florêncio