Pi Dash -- AI Agent Orchestration Platform

Website • Documentation • Community • X

Pi Dash is an open-source AI agent orchestration platform built for As Coding (asynchronous vibe coding) — a workflow where you define what needs to be built, and coding agents handle the implementation in the background. Instead of babysitting agent runs and watching terminals scroll, Pi Dash lets you focus on the work that matters: scoping tasks, reviewing results, and shipping products.

Pi Dash is evolving every day. Your suggestions, ideas, and reported bugs help us immensely. Do not hesitate to open a GitHub discussion or raise an issue. We read everything and respond to most.

🌟 Architecture

Pi Dash is composed of three major components:

Pi Dash Platform

The web-based orchestration hub where you manage projects, define tasks, and monitor agent progress. Create work items, organize them into cycles and modules, review agent output, and track analytics — all from a single dashboard. This is where you spend your time instead of watching terminals.

Pi Dash CLI & Runner Daemon

A local command-line tool and background daemon that runs on your development machine. The CLI connects to the Pi Dash platform, picks up assigned tasks, dispatches them to your configured AI agent, and reports results back. The runner daemon keeps this loop going continuously so you don't have to trigger each task manually.

AI Agent (user-provided)

Pi Dash is agent-agnostic — bring your own coding agent. Whether you use Apple Pi, Claude Code, Codex, OpenClaw, or a custom solution, Pi Dash orchestrates the workflow around it. You configure which agent the runner invokes; Pi Dash handles the rest.

🚀 Installation

Pi Dash Cloud is coming soon. For now, self-hosting is the way to go.

1. Pi Dash Platform (self-hosted)

Requirements

• Docker Engine installed and running
• Node.js version 20+ LTS version
• Python version 3.8+
• Postgres version v14
• Redis version v6.2.7
• Memory: Minimum 12 GB RAM recommended

  Running the project on a system with only 8 GB RAM may lead to setup failures or memory crashes (especially during Docker container build/start or dependency install). Use cloud environments like GitHub Codespaces or upgrade local RAM if possible.

Setup

1. Clone the repo

git clone https://github.com/The-AI-Republic/pi-dash.git [folder-name]
cd [folder-name]
chmod +x setup.sh

2. Run setup.sh

./setup.sh

setup.sh copies every .env.example to its .env counterpart (the repo root plus apps/web, apps/api, apps/space, apps/admin, apps/live), generates a unique Django SECRET_KEY and appends it to apps/api/.env, then runs pnpm install. For the default loopback-dev setup you do not need to edit any .env file manually — the .env.example defaults (localhost URLs, pi-dash database credentials, a local MinIO endpoint, etc.) work out of the box. Edit them only if you're binding to a non-default host/port or wiring in external services.

3. Start the containers

docker compose -f docker-compose-local.yml up

4. Start web apps:

pnpm dev

5. Open your browser to http://localhost:3001/god-mode/ and register yourself as instance admin
6. Open your browser to http://localhost:3000 and log in using the same credentials

Production deployment

For real self-hosted deployments (not local development), pick the path that matches how much you want to manage yourself:

• All-in-One Docker image — one container that bundles every Pi Dash service, managed by supervisord internally. Simplest path: a single docker run command. Best for demos, homelab setups, evaluation, and small teams. External Postgres / Redis / RabbitMQ / S3-compatible storage are still required.
• Docker Compose / Swarm self-hosting — the full microservices stack (6 service containers + database + queue + storage). More configuration, but gives you independent scaling and rolling updates per service. Recommended for anything beyond evaluation.
• Kubernetes / Helm — Helm chart publishing is planned but not yet shipped; see deployments/kubernetes/community/README.md.

2. Pi Dash CLI & Runner Daemon

Install the CLI on any machine where you want agents to pick up and execute tasks. Currently supported platforms:

• macOS — Apple Silicon (arm64)
• Linux — arm64 and x86_64

Intel Macs are not in the prebuilt matrix — GitHub retired the macos-13 runner image. You can still build from source on Intel macOS with Rust 1.93+.

Run the following command in your dev machine terminal:

curl --proto '=https' --tlsv1.2 -LsSf \
  https://github.com/The-AI-Republic/pi-dash/releases/latest/download/pidash-installer.sh | sh

Then register with your Pi Dash platform, install the OS service, and start the daemon:

# Register with your Pi Dash instance (grab the one-time token from the web UI)
pidash configure --url https://your-pidash-instance.com --token <ONE_TIME_CODE>

# Install as a system service (systemd on Linux, launchd on macOS)
pidash install

# Start the daemon
pidash start

The runner daemon runs in the background, polls for assigned tasks, dispatches them to your AI agent, and reports results back to the platform.

Useful commands:

Command	Description
pidash status	Print service and daemon status
pidash tui	Open interactive terminal UI to monitor the daemon
pidash doctor	Run preflight checks (agent installed, git configured, platform reachable)
pidash stop	Stop the daemon

See pidash --help for all available commands.

3. AI Agent (user-provided)

Pi Dash does not ship an AI agent — you bring your own. Ensure your chosen agent is installed and accessible on the machine running the Pi Dash CLI. Supported agents include Apple Pi, Claude Code, Codex, OpenClaw, and any custom agent that conforms to the Pi Dash agent interface.

⚙️ Built with

📝 Documentation

Explore the Pi Dash documentation to learn about features, setup, and usage.

❤️ Community

Join the conversation on GitHub Discussions, follow @ai_republic on X, or visit airepublic.com for updates. We follow a Code of conduct in all our community channels.

Feel free to ask questions, report bugs, participate in discussions, share ideas, request features, or showcase your projects. We'd love to hear from you!

🛡️ Security

If you discover a security vulnerability in Pi Dash, please report it responsibly instead of opening a public issue. See SECURITY.md for more info.

To disclose any security issues, please email us at privacy_security@airepublic.com.

🤝 Contributing

There are many ways you can contribute to Pi Dash:

• Report bugs or submit feature requests.
• Review the documentation and submit pull requests to improve it—whether it's fixing typos or adding new content.
• Show your support by upvoting popular feature requests.

Please read CONTRIBUTING.md for details on the process for submitting pull requests.

We couldn't have done this without you.

Our community contributors

Acknowledgements

Pi Dash is built on top of Plane, an open-source project management tool. We are grateful to the Plane team and its contributors for laying the foundation that made Pi Dash possible.

License

This project is licensed under the GNU Affero General Public License v3.0.