* feat: implement session restore for crashed/exited agents
Add true in-place session restore: same session ID, same worktree, same
metadata — optionally resuming the Claude Code conversation via --resume.
Core changes:
- Add TERMINAL_STATUSES, TERMINAL_ACTIVITIES, NON_RESTORABLE_STATUSES sets
and isTerminalSession/isRestorable helpers to types.ts
- Add SessionNotRestorableError and WorkspaceMissingError error classes
- Add restore() to SessionManager with 9-step flow: find metadata →
validate restorability → check/recreate workspace → get restore or
launch command → create runtime → update metadata
- Add restoredAt field to Session and SessionMetadata
Plugin extensions:
- workspace-worktree: exists() + restore() (git worktree prune + re-add)
- workspace-clone: exists() + restore() (git clone + checkout)
- scm-github: branchExists() via git rev-parse
- agent-claude-code: getRestoreCommand() finds latest JSONL session file
and builds claude --resume command
CLI + Web:
- Add `ao session restore <id>` subcommand
- Web restore API route uses sessionManager.restore() instead of spawn()
- SessionCard uses centralized TERMINAL_STATUSES/TERMINAL_ACTIVITIES
- Web types re-export core constants with sync tests
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: add "merged" to TERMINAL_STATUSES
The old inline isTerminal check included "merged" but when refactored
to use the TERMINAL_STATUSES set, "merged" was omitted. This caused
merged sessions (whose activity is not "exited") to incorrectly show
the "terminal" link and "terminate session" button.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: enrich runtime state before restore check, remove dead branchExists
- Add enrichSessionWithRuntimeState() call before isRestorable() in
restore() so crashed sessions (status "working", agent exited) are
correctly detected as terminal and eligible for restore.
- Remove dead branchExists from SCM interface and scm-github plugin
(defined but never called anywhere in the codebase).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: allow restore of crashed working sessions
Remove "working" from NON_RESTORABLE_STATUSES. The isTerminalSession()
gate already prevents restoring truly active sessions (activity is not
"exited"). This fix allows crashed agents (status "working", activity
"exited") to be restored, aligning core behavior with the UI which
already shows the restore button for this case.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: distinguish missing branch from missing restore support
Split the compound condition so workspace restore gives an accurate
error message when branch metadata is null ("branch metadata is
missing") vs when the workspace plugin lacks a restore method.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
|
||
|---|---|---|
| .. | ||
| __tests__ | ||
| src | ||
| README.md | ||
| package.json | ||
| tsconfig.build.json | ||
| tsconfig.json | ||
| vitest.config.ts | ||
README.md
@agent-orchestrator/core
Core services, types, and configuration for the Agent Orchestrator system.
What's Here
src/types.ts— All TypeScript interfaces (Runtime, Agent, Workspace, Tracker, SCM, Notifier, Terminal, Session, events)src/services/— Core services (SessionManager, LifecycleManager, PluginRegistry)src/config.ts— Configuration loading + Zod schemassrc/utils/— Shared utilities (shell escaping, metadata parsing, etc.)
Key Files
src/types.ts — The Source of Truth
Every interface the system uses is defined here. If you're working on any part of the orchestrator, start by reading this file.
Main interfaces:
Runtime— where sessions execute (tmux, docker, k8s)Agent— AI coding tool adapter (claude-code, codex, aider)Workspace— code isolation (worktree, clone)Tracker— issue tracking (GitHub Issues, Linear)SCM— PR/CI/reviews (GitHub, GitLab)Notifier— push notifications (desktop, Slack, webhook)Terminal— human interaction UI (iTerm2, web)Session— running agent instance (state, metadata, handles)OrchestratorEvent— events emitted by lifecycle managerPluginModule— what every plugin exports
src/services/session-manager.ts — Session CRUD
Handles session lifecycle:
spawn(config)— create new session (workspace + runtime + agent)list(projectId?)— list all sessionsget(sessionId)— get session detailskill(sessionId)— terminate sessioncleanup(projectId?)— kill completed/merged sessionssend(sessionId, message)— send message to agent
Data flow in spawn():
- Load project config
- Validate issue exists via
Tracker.getIssue()(if issueId provided, fails-fast if not found) - Reserve session ID
- Determine branch name
- Create workspace via
Workspace.create() - Generate prompt via
Tracker.generatePrompt() - Build launch command via
Agent.getLaunchCommand() - Create runtime session via
Runtime.create() - Run
Agent.postLaunchSetup()(optional) - Write metadata file
- Return Session object
Note: If issue validation fails (not found, auth error), spawn fails before creating any resources (no workspace, no runtime, no session ID). This prevents spawning sessions with broken issue references.
src/services/lifecycle-manager.ts — State Machine + Reactions
Polls sessions, detects state changes, triggers reactions:
State machine:
spawning → working → pr_open → ci_failed/review_pending/approved → mergeable → merged
Reactions:
ci-failed→ send fix prompt to agentchanges-requested→ send review comments to agentapproved-and-green→ notify human (or auto-merge)agent-stuck→ notify human
Polling loop:
- For each session: check agent activity state (
Agent.getActivityState()) - If PR exists: check CI status (
SCM.getCISummary()), review state (SCM.getReviewDecision()) - Update session status based on state
- Trigger reactions if state changed
- Emit events
src/services/plugin-registry.ts — Plugin Discovery + Loading
Loads plugins and provides access to them:
register(plugin, config?)— register a plugin instanceget<T>(slot, name)— get plugin by slot + namelist(slot)— list all plugins for a slotloadBuiltins(config?)— load built-in plugins (runtime-tmux, agent-claude-code, etc.)loadFromConfig(config)— load plugins from config (npm packages, local paths)
Built-in plugins (loaded by default):
- runtime-tmux, runtime-process
- agent-claude-code, agent-codex, agent-aider, agent-opencode
- workspace-worktree, workspace-clone
- tracker-github, tracker-linear
- scm-github
- notifier-desktop, notifier-slack, notifier-composio, notifier-webhook
- terminal-iterm2, terminal-web
src/config.ts — Configuration Loading
Loads and validates agent-orchestrator.yaml:
Main config sections:
dataDir— where session metadata lives (~/.agent-orchestrator)worktreeDir— where workspaces are created (~/.worktrees)port— web dashboard port (default 3000)defaults— default plugins (runtime, agent, workspace, notifiers)projects— per-project config (repo, path, branch, symlinks, reactions, agentRules)notifiers— notification channel config (Slack webhooks, etc.)notificationRouting— which notifiers get which priority eventsreactions— auto-response config (ci-failed, changes-requested, approved-and-green, etc.)
Zod schemas validate all config at load time.
Common Tasks
Adding a Field to Session
- Edit
src/types.ts→Sessioninterface - Edit
src/services/session-manager.ts→ initialize field inspawn() - Rebuild:
pnpm --filter @agent-orchestrator/core build
Adding an Event Type
- Edit
src/types.ts→EventTypeunion - Emit the event:
eventEmitter.emit()in relevant service - Add reaction handler (optional):
src/services/lifecycle-manager.ts
Adding a Reaction
- Edit
src/services/lifecycle-manager.ts→ add handler function - Wire it up in the polling loop
- Add config schema in
src/config.tsif new reaction type
Testing
# Run all core tests
pnpm --filter @agent-orchestrator/core test
# Run in watch mode
pnpm --filter @agent-orchestrator/core test -- --watch
# Run specific test
pnpm --filter @agent-orchestrator/core test -- session-manager.test.ts
Tests are in src/__tests__/:
session-manager.test.ts— session CRUD, spawn, cleanuplifecycle-manager.test.ts— state machine, reactionsplugin-registry.test.ts— plugin loading, resolutiontmux.test.ts— tmux utility functions (not a plugin test)prompt-builder.test.ts— prompt generation utilities
Building
# Build core
pnpm --filter @agent-orchestrator/core build
# Typecheck
pnpm --filter @agent-orchestrator/core typecheck
This package is a dependency of all other packages. Build it first if working on the codebase.
Architecture Notes
Why flat metadata files?
- Debuggability:
cat ~/.agent-orchestrator/my-app-3shows full state - No database dependency (survives crashes, easy to inspect)
- Backwards-compatible with bash script orchestrator
Why polling instead of webhooks?
- Simpler (no webhook setup, no ngrok for local dev)
- Works offline (CI/review state is fetched, not pushed)
- Survives orchestrator restarts (no missed events)
Why plugin slots?
- Swappability: use tmux locally, docker in CI, k8s in prod
- Testability: mock plugins for tests
- Extensibility: users can add custom plugins (e.g., company-specific notifier)