KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #dsh-plugin #developer-tools #ai-agent #claude-code #codex #typescript #ai-agents #deepseek-harness #mcp #cli #rust

nakama

★ 214 Open GitHub ↗

It's like Hermes Agent & OpenClaw but designed to work nicely with teams.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Nakama

[Documentation](https://ahmadrosid.github.io/nakama/) · [Demo](https://demo.getnakama.cloud) · [Managed hosting](https://getnakama.cloud/)

> Your next hire will still be human.
> Nakama runs the agents. Your team keeps the important work.

Nakama is an AI agent platform for your team. You chat in a dashboard that feels like ChatGPT. Behind that, each agent can use tools, skills, memory, and channels (Telegram, WhatsApp, Discord) — with its own identity and permissions.

It was built for teams from day one: one server, many orgs, invites and roles, agents isolated from each other. Set it up in the browser — demo, managed cloud, or one Docker container.

Coming from [OpenClaw](https://github.com/openclaw/openclaw) or [Hermes Agent](https://github.com/nousresearch/hermes-agent)? Same class of capability, without the setup maze, and not limited to one operator on one machine.

Open [ARCHITECTURE.md](./ARCHITECTURE.md) for the system design.
Open the [docs site](https://ahmadrosid.github.io/nakama/) for the full guide.

## Quick start

### Try the demo

Open the live demo at [https://demo.getnakama.cloud](https://demo.getnakama.cloud).

- Username: `demo@getnakama.cloud`
- Password: `demo1234`

### Managed hosting

Use [Nakama Cloud](https://getnakama.cloud/) to try Nakama with the least work.

1. Create an account.
2. Provision an instance.
3. Complete the first-time setup wizard in the browser.

You do not need Bun, Docker, or a VPS.

### Run locally

You need [Bun](https://bun.sh).

```bash
# Install dependencies
bun install

# Start the web (starts the server automatically if needed)
bun run dev:web
```

Open the web dashboard: http://localhost:3000

Or start the server alone:

```bash
bun run dev:server
```

### Docker

You can also run Nakama with Docker.

**Prebuilt image (fastest):**

```bash
# Pull and run the latest image
docker pull ghcr.io/ahmadrosid/nakama:latest
docker run -d -p 4310:4310 -v nakama-data:/nakama/data --name nakama ghcr.io/ahmadrosid/nakama:latest
```

**Build from source:**

```bash
./scripts/docker-build-run.sh
```

**Fresh start:**

```bash
./scripts/docker-destroy.sh
./scripts/docker-build-run.sh
```

Open the dashboard at http://localhost:4310.

The image carries [omni](https://github.com/fajarhide/omni), which shortens `bash`
and `read_file` output before it reaches the model. Turn it on under Integrations,
or with `NAKAMA_OMNI=1`. Build with `--build-arg OMNI_VERSION=` to leave the binary
out; the server then downloads it, checksum verified, the first time the toggle is
switched on. Set `NAKAMA_OMNI_AUTO_INSTALL=0` to forbid that download.

### Integrations

Nakama connects to **Telegram**, **WhatsApp**, and **Composio**.
With Composio, you can connect to more than 1,000 external apps.
Enable them in the web app under **Integrations**.

On the first run, the server asks for a provider and an API key if none is configured.
The server saves settings to `~/.nakama/config.ini`.

The server listens on `http://127.0.0.1:4310` by default.
Interactive API docs are at `http://127.0.0.1:4310/docs`.

## License

MIT

nakama

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README