You're still babysitting AI agents

You have 3 AI assistants open. Claude is implementing auth in one terminal. Codex is writing tests in another. A shell script runs migrations somewhere else.

You're the human router — switching tabs, copy-pasting context, manually tracking who's doing what, restarting crashed agents at 2am.

You're not the founder. You're the intern.

What if you had an entire AI department?

$ orch org deploy startup-mvp --goal "Implement user auth with OAuth2"

  ✓ Deployed team "platform" — 5 agents
    CTO (claude)       → Decomposing goal into tasks...
    Backend A (claude)  → Waiting for tasks
    Backend B (codex)   → Waiting for tasks
    QA (codex)          → Waiting for tasks
    Reviewer (claude)   → Waiting for reviews

  ✓ CTO created 6 tasks from goal

$ orch run --all --watch

  22:03  ▶ Backend A    → "Implement OAuth2 flow"         [feature/oauth]
  22:03  ▶ Backend B    → "JWT token service"              [feature/jwt]
  22:03  ▶ QA           → waiting for implementations...
  22:15  ✓ Backend B    DONE  (12m · 4,200 tokens)
  22:15  ▶ QA           → "Test JWT service"               [test/jwt]
  22:22  ✓ Backend A    DONE  (19m · 8,100 tokens)
  22:24  ↻ QA           RETRY  attempt 2/3
  22:28  ✓ QA           DONE  (6m · 2,800 tokens)
  22:29  ▶ Reviewer     → "Review OAuth2 implementation"
  22:33  ✓ Reviewer     DONE  → all tasks in review

  → You went to sleep at 22:05.
  → You wake up to 6 tasks in review. Approve. Merge. Ship.

One goal. Five agents. Six PRs. Zero tab-switching. $4.20 in tokens.

Launch your AI company in 30 seconds

npm install -g @oxgeneral/orch

cd ~/your-project && orch

That's it. ORCH auto-initializes, opens the TUI command center. Add agents and tasks right from there.

Or deploy a pre-built department:

orch org deploy startup-mvp --goal "Build invoicing SaaS with Stripe"
orch run --all --watch

Requirements: Node.js >= 20. That's it. No database. No cloud. No Docker.

Your AI company org chart

CTO — strategic decomposition

Set a high-level goal. Your CTO agent decomposes it into concrete tasks, assigns priorities, and delegates to the right departments. You set strategy — AI executes.

Engineering Department — parallel execution

Backend A, Backend B, Frontend — each agent gets its own git worktree (isolated branch). They work in parallel without file conflicts. Failed? Auto-retry with exponential backoff. Stalled? Zombie detection kills and re-queues.

QA Department — automated verification

QA agents pick up completed work, run tests, validate contracts. Reject with feedback → task goes back to engineering with your notes. The loop closes automatically.

Code Review — mandatory quality gate

Nothing touches main until reviewed. Every task flows through the state machine:

todo → in_progress → review → done
                    ↘ retrying → in_progress
                    ↘ failed

Every transition validated. No task gets lost. No code merges without approval.

Inter-department communication

Agents talk to each other — direct messages, team broadcasts, shared context store. Backend finishes auth module → sends message to QA → QA starts testing. No copy-paste. No manual routing.

Not just engineering

ORCH orchestrates any process — not just code. The shell adapter runs any CLI tool, which means any workflow becomes a zero-human department:

| Department | Agents | What they do | |-----------|--------|-------------| | Engineering | Claude, Codex, Cursor | Write code, fix bugs, refactor | | Editorial | Claude (writer), Claude (editor), Shell (grammarly) | Write articles, edit, check grammar, publish | | Sales Ops | Shell (CRM scripts), Claude (copywriter), Shell (email sender) | Generate leads, write sequences, send outreach | | Analytics | Shell (pandas, duckdb), Claude (analyst), Shell (matplotlib) | Clean data, compute KPIs, generate reports | | Content Factory | Claude (strategist), Claude (writer x2), Claude (SEO) | Plan content calendar, write posts, optimize | | Security | Shell (Semgrep, Trivy, Gitleaks), Claude (hunter) | Scan code, correlate findings, auto-fix | | DevOps | Shell (terraform, kubectl), Claude (architect) | Plan infra changes, apply, verify |

Every department gets the same superpowers: state machine governance, retry, messaging, isolation, review gate.

Why founders choose ORCH

You're the CEO, not the babysitter

Set goals, not tasks. Your CTO agent decomposes strategy into work. Backend builds. QA tests. Reviewer checks. You approve. That's the workflow.

Real departments, not just agents

Teams with leads, shared task pools, auto-claiming. CTO, Backend, QA, Reviewer — organized like a real org chart. Broadcasts, messaging, shared context.

Runs while you sleep

Auto-retry, stall detection, exponential backoff. Your agents don't stop when one fails — ORCH restarts, re-queues, and keeps going. Deploy at night, review in the morning.

Main branch is sacred

Every agent works on its own git branch. Scope overlap detection prevents conflicts before they happen. Mandatory review step. Nothing merges until you approve.

30 seconds to a full department

orch org deploy startup-mvp — CTO, backend x2, QA, reviewer. Team created. Goal set. Agents ready. One command.

Free. Yours. Forever.

MIT license. Your API keys. Your machine. State in .orchestry/ — plain YAML, JSON, JSONL files you can read, edit, commit. Zero cloud dependencies.

Works with every tool — AI or not

| Adapter | What it runs | Use as | |---------|-------------|--------| | claude | Claude Code CLI | CTO, Backend, Reviewer, Writer, Analyst | | opencode | OpenCode (Gemini, DeepSeek via OpenRouter) | Backend, Frontend, Writer | | codex | OpenAI Codex CLI | Backend, QA | | cursor | Cursor Agent CLI (headless) | Frontend, Backend | | shell | Anything: npm test, python bot.py, Semgrep, curl, CRM scripts, email senders, data pipelines | QA, Security, DevOps, Sales, Analytics, Content — any role |

The shell adapter is the key: if it runs in a terminal, it's an employee. This is how ORCH goes beyond engineering — any CLI tool becomes part of your zero-human company.

Pre-built AI companies

Deploy a full department with one command:

Engineering

| Template | Agents | What it does | |----------|--------|-------------| | startup-mvp | CTO, Backend x2, Frontend, QA, Reviewer | Ship an MVP in 48 hours | | pr-review-corp | Security, Performance, Style, QA, CTO | Automated review for every PR | | migration-squad | CTO, Migrator x3, QA, Reviewer | JS-to-TS migration over a weekend | | security-dept | Semgrep, Trivy, Gitleaks, Hunter, Reviewer | Multi-layer security audit | | test-factory | Coverage, Backend x2, QA x2, Reviewer | Coverage from 40% to 80% overnight | | bugfix-dept | Triager, Fixer x3, QA, Reviewer | 100 issues to 0 in a week |

Non-Engineering

| Template | Agents | What it does | |----------|--------|-------------| | content-agency | Strategist, Writer x2, Editor, SEO | Content factory: plan, write, edit, optimize | | data-lab | Shell (pandas, duckdb, matplotlib), Analyst | 3 CSVs → executive report by morning | | sales-machine | Researcher, Copywriter, Shell (email), QA | Lead gen → personalized sequences → send | | docs-team | Architect, Writer x2, Editor, Reviewer | Technical docs from codebase analysis |

orch org list                                    # See all companies
orch org deploy startup-mvp                      # Deploy the default
orch org deploy startup-mvp --goal "Build X"     # Deploy with a goal
orch org export my-company                       # Save your setup as template

Full CLI reference

# Setup
orch init                          # Initialize project
orch doctor                        # System diagnostics

# Departments & Agents
orch agent add <name> --adapter claude --role "CTO — decomposes goals"
orch agent list                    # Status of all agents
orch agent disable/enable <id>     # Toggle availability

# Organization Templates
orch org list                      # List available companies
orch org deploy <template>         # Deploy a full department
orch org deploy <template> --goal "..." # Deploy with a goal
orch org export <name>             # Export current setup

# Tasks
orch task add "Title" -p 1         # Create task (priority 1-4)
orch task list                     # List all tasks
orch task assign <task> <agent>    # Manual assignment
orch task cancel <task>            # Cancel running task

# Teams (Departments)
orch team create <name> --lead <agent-id>
orch team join <team-id> <agent-id>
orch team add-task <team-id> <task-id>
orch team disband <id>

# Goals (Strategy)
orch goal add "Title" --description "..."
orch goal list
orch goal status <id> achieved

# Communication
orch msg send <agent-id> "message"              # Direct message
orch msg broadcast "message" --team <team-id>   # Team broadcast
orch context set <key> <value>                  # Shared context

# Execution
orch run --all --watch             # Launch all agents
orch run <task-id>                 # Run single task

# Monitoring
orch status                        # Quick overview
orch logs <run-id>                 # View run logs
orch tui                           # Command center (TUI)

# Config
orch config edit                   # Open in $EDITOR
orch update                        # Check for updates

Aliases: orchestry, orch, ao

Architecture

ORCH is an engine first, CLI second. The core has zero dependencies on CLI/TUI layers — you can import @oxgeneral/orch as a library and build your own interface.

src/
├── domain/           # Models, state machine, errors
├── application/      # Orchestrator engine, services, event bus
├── infrastructure/
│   ├── adapters/     # Claude, OpenCode, Codex, Cursor, Shell
│   ├── storage/      # File-based (YAML/JSON/JSONL)
│   ├── process/      # PID management, graceful kill
│   ├── template/     # LiquidJS prompt templates
│   └── workspace/    # Git worktree isolation
├── cli/              # Commander.js commands
└── tui/              # Ink + React command center

Development

npm run dev            # Run via tsx
npm run build          # Build ESM + DTS
npm test               # 1493 tests via Vitest
npm run typecheck      # Strict TypeScript

Community

If ORCH saves you time — Star it on GitHub — it helps other founders find the project.

• Open an issue if something breaks or could be better
• Submit a PR — see CONTRIBUTING.md

Used by / Early Adopters

Running a zero-human company with ORCH? Let us know — we'd love to feature you here.

Be the first to launch your AI company here.

FAQ

Do I need a team to use ORCH?

No. Solo founders are the primary users. You + 2 agents is already a zero-human company. ORCH gives you auto-retry, state machine, dashboard, and token tracking even with a single agent. Start with one, scale to departments.

Will agents mess up my codebase?

No. Every agent works in an isolated git worktree on its own branch. Nothing touches main until you explicitly approve. Mandatory review step in the state machine. Scope overlap detection prevents conflicts before they happen.

Is this only for engineering?

No. The shell adapter runs any CLI tool — which means ORCH orchestrates any process. Engineering, editorial, sales, analytics, security, DevOps. If it runs in a terminal, it's an employee. Deploy content-agency, data-lab, or sales-machine with one command.

How much does it cost?

ORCH is free forever (MIT). You pay only for the AI APIs you already use (Claude, Codex, etc.). The TUI shows token counts per agent per run in real-time — no surprise bills.

What AI tools does it support?

Five adapters: Claude Code, OpenCode (Gemini, DeepSeek, any OpenRouter model), Codex, Cursor, and Shell (any CLI tool). Your API keys, your tools — ORCH coordinates them.

Is there a cloud component?

None. Zero cloud. All state in .orchestry/ — plain YAML, JSON, JSONL files. No signup, no account, no data leaves your machine.

How is this different from Paperclip?

Paperclip needs PostgreSQL, a web server, and cloud setup. ORCH needs npm install. Same vision — zero-human companies — but ORCH is the hacker's version: terminal-first, file-based, zero infrastructure, MIT licensed.

License

MIT — build whatever you want.