LlamaBot

The open-source AI coding agent that rapidly builds MVPs, prototypes, and internal tools.

---

✨ What is LlamaBot?

LlamaBot is an AI coding agent that generates working prototypes, embeds AI directly into the app, and runs real workflows — letting you move from idea to production in record time.

It works across the full spectrum of users:

• Non-technical founders who want to build without code.
• Product teams who need to spin up prototypes fast.
• Engineers who want an AI teammate to automate workflows inside production apps.

Unlike typical codegen tools, LlamaBot doesn’t just write snippets — it can embed directly in your app and run real workflows. This makes it ideal for collaborative software building: founders guide the vision, engineers stay in control, and AI fills the gap.

---

LlamaBot is built for moving ideas fast:

• 🚀 Prototype an AI MVP in a weekend — landing pages, user flows, backend logic, all scaffolded by AI.
• 🧪 Experiment with workflows — test reporting, notifications, automations directly inside your app.
• 👥 Collaborate with teammates — Bridge the gap between non-technical founders and engineering teams.

---

🚀 Quick Start (<5 Minutes)

Requires:

• Docker Compose
• OpenAI API Key

Run the install script remotely (no Github clone)

# Only requirement: docker compose + your OpenAI key
curl -fsSL "https://raw.githubusercontent.com/KodyKendall/LlamaBot/refs/heads/main/bin/install_llamabot_local.sh" -o install_llamabot_local.sh && bash install_llamabot_local.sh

Open your browser:

http://localhost:8000/chat

🚀 Dev Start (5-10 Minutes)

Clone repo & run install script locally

git clone https://github.com/kodykendall/LlamaBot
cd LlamaBot
bash bin/install_llamabot_local.sh

🔌 Embed in an Existing App (Rails first, others coming)

Today, Rails is the primary supported framework. With the llama_bot_rails gem, you can use LlamaBot to:

• Call existing ActiveRecord models
• Trigger your real services, jobs, and routes
• Automate workflows with natural language

Example use cases:

• Refund a user and send SMS
• Generate a weekly revenue report
• Queue 100 Sidekiq jobs from chat

Future adapters: Django, Laravel, Node.js.

Not a developer but want to build something with LlamaBot? Join the Discord or reach out directly — we’d love to collaborate on real-world MVPs and prototypes.

---

🧠 Under the Hood

• Built on LangGraph for multi-step agent workflows
• FastAPI + WebSockets for streaming real-time responses
• Scoped memory per conversation/session
• Can call external APIs or internal app methods via whitelisted routes

---

📦 Project Structure

LlamaBot/
├── app/
│   ├── main.py            # FastAPI app with WebSocket + API routes
│   ├── chat.html          # Chat interface UI
│   ├── page.html          # Agent scratchpad to display visual UI to user, show results, etc. (Such as the Agent TODO list, etc.)
│   ├── agents/            # LangGraph agent logic
│   ├── main.py            # FastAPI app with WebSocket + API routes
│   ├── chat.html          # Chat interface UI
│   └── ...                # Utility code, workflows, memory, etc.
├── bin/
│   ├── install_llamabot_local.sh # local dev install script
│   └── install_llamabot_prod.sh  # production deployment script
├── docs/
│   └── dev_logs/
│      ├── ...
│      ├── v0.1.7
│      └── v0.2.0
├── Dockerfile             # Run backend anywhere
├── requirements.txt       # Python dependencies
└── README.md

---

🤝 Contributing

We welcome PRs, issues, and ideas! Jump into Discord to collaborate.

📜 License

LlamaBot is AGPLv3 open source. For commercial licensing, contact kody@llamapress.ai.

Made with ❤️ in San Francisco — inspired by the next wave of AI-powered software.