Overstory
Multi-agent orchestration for AI coding agents.
No longer maintained. Overstory is no longer actively maintained and this repository is archived (read-only) — I am not reviewing issues or merging pull requests here anymore. Thank you to everyone who contributed. The code stays MIT-licensed, so feel free to fork and carry it forward. Active development has moved to Warren, a self-hostable control plane for sandboxed cloud agents and overstory's spiritual successor — start there for anything new.
Overstory turns a single coding session into a multi-agent team by spawning worker agents in isolated git worktrees, coordinating them through a custom SQLite mail system, and merging their work back with tiered conflict resolution. New projects spawn Claude agents headless and surface them through a web UI (ov serve); tmux attach is the opt-in escape hatch for live steering. A pluggable AgentRuntime interface lets you swap between 11 runtimes — Claude Code, Pi, Gemini CLI, Aider, Goose, Amp, or your own adapter.
Warning: Agent swarms are not a universal solution. Do not deploy Overstory without understanding the risks of multi-agent orchestration — compounding error rates, cost amplification, debugging complexity, and merge conflicts are the normal case, not edge cases. Read STEELMAN.md for a full risk analysis and the Agentic Engineering Book (web version) before using this tool in production.
Install
Requires Bun v1.0+ and git. tmux is optional — only needed if you want to spawn workers with --no-headless or attach to a coordinator/worker pane directly. At least one supported agent runtime must be installed:
bun install -g @os-eco/overstory-cli
Or try without installing:
npx @os-eco/overstory-cli --help
Development
git clone https://github.com/jayminwest/overstory.git
cd overstory
bun install
bun link # Makes 'ov' available globally
bun test # Run all tests
bun run lint # Biome check
bun run typecheck # tsc --noEmit
Quick Start
# Initialize overstory in your project
cd your-project
ov init
# Install hooks into .claude/settings.local.json
ov hooks install
# Start a coordinator (persistent orchestrator)
ov coordinator start
# Open the web UI — primary operator surface for the swarm
ov serve # then open http://localhost:7321
ov serve is where you watch the fleet, read the mail bus, and inspect
per-agent timelines. New projects spawn Claude workers headless by default,
so the UI sees them with full structured-event fidelity.
Other common commands:
# Spawn an individual worker agent (coordinator usually does this for you)
ov sling <task-id> --capability builder --name my-builder
# Force a worker into tmux when you want to attach mid-session
ov sling <task-id> --capability builder --name my-builder --no-headless
tmux attach -t ov-my-builder
# Inspect state from the CLI (also visible in the UI)
ov status
ov dashboard # live TUI alternative to the web UI
ov mail check --inject
ov nudge <agent-name> # send a follow-up to a stalled agent
Commands
Every command supports --json where noted. Global flags: -q/--quiet, --timing, --project <path>. ANSI colors respect NO_COLOR.
Core Workflow
| Command |
Description |
ov init |
Initialize .overstory/ and bootstrap os-eco tools (--yes, --name, --tools, --skip-mulch, --skip-seeds, --skip-canopy, --skip-onboard, --json) |
ov sling <task-id> |
Spawn a worker agent (--capability, --name, --spec, --files, --parent, --depth, --skip-scout, --skip-review, --max-agents, --dispatch-max-agents, --skip-task-check, --no-scout-check, --runtime, --base-branch, --profile, --headless, --no-headless, --recover, --json) |
ov stop <agent-name> |
Terminate a running agent (--clean-worktree, --json) |
ov prime |
Load context for orchestrator/agent (--agent, --compact) |
ov spec write <task-id> |
Write a task specification (--body) |
ov discover |
Discover a brownfield codebase via coordinator-driven scout swarm (--skip, --name, --attach, --watchdog, --json) |
ov update |
Refresh .overstory/ managed files from installed package (--agents, --manifest, --hooks, --dry-run, --json) |
Coordination
| Command |
Description |
ov coordinator start |
Start persistent coordinator agent (--attach/--no-attach, --watchdog, --monitor, --profile) |
ov coordinator stop |
Stop coordinator |
ov coordinator status |
Show coordinator state |
ov coordinator send |
Fire-and-forget message to coordinator (--subject) |
ov coordinator ask |
Synchronous request/response to coordinator (--subject, --timeout) |
ov coordinator output |
Show recent coordinator output (--lines) |
ov coordinator check-complete |
Evaluate exit triggers, return completion status |
ov orchestrator start |
Start multi-repo orchestrator agent (--attach/--no-attach, --watchdog, --profile) |
ov orchestrator stop |
Stop orchestrator |
ov orchestrator status |
Show orchestrator state |
ov orchestrator send |
Fire-and-forget message to orchestrator (--subject) |
ov orchestrator ask |
Synchronous request/response to orchestrator (--subject, --timeout) |
ov orchestrator output |
Show recent orchestrator output (--lines) |
ov supervisor start |
[DEPRECATED] Start per-project supervisor agent |
ov supervisor stop |
[DEPRECATED] Stop supervisor |
ov supervisor status |
[DEPRECATED] Show supervisor state |
Messaging
| Command |
Description |
ov mail send |
Send a message (--to, --subject, --body, --type, --priority) |
ov mail check |
Check inbox — unread messages (--agent, --inject, --debounce, --json) |
ov mail list |
List messages with filters (--from, --to, --unread) |
ov mail read <id> |
Mark message as read |
ov mail reply <id> |
Reply in same thread (--body) |
ov nudge <agent> [message] |
Send a text nudge to an agent (--from, --force, --json) |
Task Groups
| Command |
Description |
ov group create <name> |
Create a task group for batch tracking |
ov group status <name> |
Show group progress |
ov group add <name> <issue-id> |
Add issue to group |
ov group list |
List all groups |
Merge
| Command |
Description |
ov merge |
Merge agent branches into canonical (--branch, --all, --into, --dry-run, --json) |
Observability
| Command |
Description |
ov status |
Show all active agents, worktrees, tracker state (--json, --verbose, --all) |
ov dashboard |
Live TUI dashboard for agent monitoring (--interval, --all) |
ov inspect <agent> |
Deep per-agent inspection (--follow, --interval, --no-tmux, --limit, --json) |
ov trace |
View agent/task timeline (--agent, --run, --since, --until, --limit, --json) |
ov errors |
Aggregated error view across agents (--agent, --run, --since, --until, --limit, --json) |
ov replay |
Interleaved chronological replay (--run, --agent, --since, --until, --limit, --json) |
ov feed |
Unified real-time event stream (--follow, --interval, --agent, --run, --json) |
ov logs |
Query NDJSON logs across agents (--agent, --level, --since, --until, --follow, --json) |
ov costs |
Token/cost analysis and breakdown (--live, --self, --agent, --run, --bead, --by-capability, --last, --json) |
ov metrics |
Show session metrics (--last, --json) |
ov run list |
List orchestration runs (--last, --json) |
ov run show <id> |
Show run details |
ov run complete |
Mark current run as completed |
Infrastructure
| Command |
Description |
ov hooks install |
Install orchestrator hooks to .claude/settings.local.json (--force) |
ov hooks uninstall |
Remove orchestrator hooks |
ov hooks status |
Check if hooks are installed |
ov worktree list |
List worktrees with status |
ov worktree clean |
Remove completed worktrees (--completed, --all, --force) |
ov watch |
Start watchdog daemon — Tier 0 (--interval, --background) |
ov monitor start |
Start Tier 2 monitor agent |
ov monitor stop |
Stop monitor agent |
ov monitor status |
Show monitor state |
ov log <event> |
Log a hook event (--agent) |
ov clean |
Clean up worktrees, sessions, artifacts (--completed, --all, --run) |
ov doctor |
Run health checks on overstory setup — 13 categories (--category, --fix, --json) |
ov serve |
HTTP + WebSocket surface for the web UI (--port, --host, --json) |
ov ecosystem |
Show os-eco tool versions and health (--json) |
ov upgrade |
Upgrade overstory to latest npm version (--check, --all, --json) |
ov agents discover |
Discover agents by capability/state/parent (--capability, --state, --parent, --json) |
ov completions <shell> |
Generate shell completions (bash, zsh, fish) |
Architecture
Overstory uses instruction overlays and tool-call guards to turn agent sessions into orchestrated workers. Each agent runs in an isolated git worktree; new projects spawn Claude workers as headless subprocesses (stream-json over stdout) and surface them through ov serve's web UI, with tmux available as an opt-in for live attach. Inter-agent messaging is handled by a custom SQLite mail system (WAL mode, ~1-5ms per query) with typed protocol messages and broadcast support. A FIFO merge queue with 4-tier conflict resolution merges agent branches back to canonical. A tiered watchdog system (Tier 0 mechanical daemon, Tier 1 AI-assisted triage, Tier 2 monitor agent) ensures fleet health. See CLAUDE.md for full technical details.
Runtime Adapters
Overstory is runtime-agnostic. The AgentRuntime interface (src/runtimes/types.ts) defines the contract — each adapter handles spawning, config deployment, guard enforcement, readiness detection, and transcript parsing for its runtime. Set the default in config.yaml or override per-agent with ov sling --runtime <name>.
Claude Code agents can run in headless mode (the default for new projects — -p --output-format stream-json subprocess, NDJSON events parsed by ClaudeRuntime.parseEvents, surfaced through ov serve's web UI) or tmux mode (escape hatch for live attach — operator can tmux attach to watch and steer mid-session). ov init writes runtime.claudeHeadlessByDefault: true for new projects; legacy projects upgrading from earlier overstory versions keep tmux until they edit config. Override per-spawn with ov sling --no-headless (force tmux) or --headless (force headless). Sapling is statically headless; Pi, Codex, and Cursor have no buildDirectSpawn and reject --headless.
| Runtime |
CLI |
Guard Mechanism |
Stability |
| Claude Code |
claude |
settings.local.json hooks |
Stable |
| Sapling |
sp |
.sapling/guards.json |
Stable |
| Pi |
pi |
.pi/extensions/ guard extension |
Experimental |
| Copilot |
copilot |
(none — --allow-all-tools) |
Experimental |
| Cursor |
agent |
(none — --yolo) |
Experimental |
| Codex |
codex |
OS-level sandbox (Seatbelt/Landlock) |
Experimental |
| Gemini |
gemini |
--sandbox flag |
Experimental |
| Aider |
aider |
(none — --yes-always) |
Experimental |
| Goose |
goose |
Profile-based permissions |
Experimental |
| Amp |
amp |
Built-in approval system |
Experimental |
| OpenCode |
opencode |
(none) |
Experimental |
How It Works
Instruction overlays + tool-call guards + the ov CLI turn your coding session into a multi-agent orchestrator. A persistent coordinator agent manages task decomposition and dispatch, while a mechanical watchdog daemon monitors agent health in the background.
Orchestrator (multi-repo coordinator of coordinators)
--> Coordinator (persistent orchestrator at project root)
--> Supervisor / Lead (team lead, depth 1)
--> Workers: Scout, Builder, Reviewer, Merger (depth 2)
Agent Types
| Agent |
Role |
Access |
| Orchestrator |
Multi-repo coordinator of coordinators — dispatches coordinators per sub-repo |
Read-only |
| Coordinator |
Persistent orchestrator — decomposes objectives, dispatches agents, tracks task groups |
Read-only |
| Supervisor |
Per-project team lead — manages worker lifecycle, handles nudge/escalation |
Read-only |
| Scout |
Read-only exploration and research |
Read-only |
| Builder |
Implementation and code changes |
Read-write |
| Reviewer |
Validation and code review |
Read-only |
| Lead |
Team coordination, can spawn sub-workers |
Read-write |
| Merger |
Branch merge specialist |
Read-write |
| Monitor |
Tier 2 continuous fleet patrol — ongoing health monitoring |
Read-only |
Key Architecture
- Agent Definitions: Two-layer system — base
.md files define the HOW (workflow), per-task overlays define the WHAT (task scope). Base definition content is injected into spawned agent overlays automatically.
- Messaging: Custom SQLite mail system with typed protocol — 8 message types (
worker_done, merge_ready, dispatch, escalation, etc.) for structured agent coordination, plus broadcast messaging with group addresses (@all, @builders, etc.)
- Worktrees: Each agent gets an isolated git worktree — no file conflicts between agents
- Merge: FIFO merge queue (SQLite-backed) with 4-tier conflict resolution
- Watchdog: Tiered health monitoring — Tier 0 mechanical daemon (tmux/pid liveness), Tier 1 AI-assisted failure triage, Tier 2 monitor agent for continuous fleet patrol
- Tool Enforcement: Runtime-specific guards (hooks for Claude Code, extensions for Pi) mechanically block file modifications for non-implementation agents and dangerous git operations for all agents
- Task Groups: Batch coordination with auto-close when all member issues complete
- Session Lifecycle: Checkpoint save/restore for compaction survivability, handoff orchestration for crash recovery
- Token Instrumentation: Session metrics extracted from runtime transcript files (JSONL)
Project Structure
overstory/
src/
index.ts CLI entry point (Commander.js program)
types.ts Shared types and interfaces
config.ts Config loader + validation
errors.ts Custom error types
json.ts Standardized JSON envelope helpers
commands/ One file per CLI subcommand (38 commands)
agents.ts Agent discovery and querying
coordinator.ts Persistent orchestrator lifecycle
supervisor.ts Team lead management [DEPRECATED]
dashboard.ts Live TUI dashboard (ANSI via Chalk)
hooks.ts Orchestrator hooks management
sling.ts Agent spawning
group.ts Task group batch tracking
nudge.ts Agent nudging
mail.ts Inter-agent messaging
monitor.ts Tier 2 monitor management
merge.ts Branch merging
status.ts Fleet status overview
prime.ts Context priming
init.ts Project initialization
worktree.ts Worktree management
watch.ts Watchdog daemon
log.ts Hook event logging
logs.ts NDJSON log query
feed.ts Unified real-time event stream
run.ts Orchestration run lifecycle
trace.ts Agent/task timeline viewing
clean.ts Worktree/session cleanup
doctor.ts Health check runner (13 check modules)
inspect.ts Deep per-agent inspection
spec.ts Task spec management
errors.ts Aggregated error view
replay.ts Interleaved event replay
stop.ts Agent termination
costs.ts Token/cost analysis
metrics.ts Session metrics
ecosystem.ts os-eco tool dashboard
update.ts Refresh managed files
upgrade.ts npm version upgrades
discover.ts Brownfield codebase discovery via coordinator-driven scout swarm
orchestrator.ts Multi-repo coordination (PersistentAgentSpec)
completions.ts Shell completion generation (bash/zsh/fish)
serve.ts HTTP + WebSocket surface for the web UI
serve/ REST handlers, WebSocket broadcaster, static SPA fallback
canopy/
client.ts Canopy client (prompt rendering, listing, emission)
agents/ Agent lifecycle management
manifest.ts Agent registry (load + query)
overlay.ts Dynamic CLAUDE.md overlay generator
identity.ts Persistent agent identity (CVs)
checkpoint.ts Session checkpoint save/restore
lifecycle.ts Handoff orchestration
hooks-deployer.ts Deploy hooks + tool enforcement
copilot-hooks-deployer.ts Deploy hooks config to Copilot worktrees
guard-rules.ts Shared guard constants (tool lists, bash patterns)
mail-poll-detect.ts Bash mail-poll pattern detector (runtime backstop)
scope-detect.ts Soft FILE_SCOPE violation detection (builder/merger)
worktree/ Git worktree + tmux management
mail/ SQLite mail system (typed protocol, broadcast)
merge/ FIFO queue + conflict resolution + sentinel-file lock + dry-run prediction
watchdog/ Tiered health monitoring (daemon, triage, health)
logg
Overstory 把单个编程会话变成一个多智能体团队：它在隔离的 git worktree 中生成 worker 智能体，通过一个自定义的 SQLite 邮件系统协调它们，并以分层冲突解决机制把它们的成果合并回来。