rowboat/README.md

Claude Code for Everything Else

- ✨ **Create background agents with full shell access** - E.g. "Generate a NotebookLM-style podcast from my saved articles every morning" - 🔧 **Connect any MCP server to add capabilities** - Add MCP servers and RowboatX handles the integration - 🎯 **Control agents with standard Unix commands** - RowboatX uses bash to spawn, monitor, and manage all background agents

With full terminal access and filesystem-as-state architecture, RowboatX gives you complete control over powerful AI automations!

Quick start

1. Set your OpenAI key

   export OPENAI_API_KEY=your-openai-api-key  
   

2. Clone the repository and start Rowboat (requires Docker)

   git clone git@github.com:rowboatlabs/rowboat.git
   cd rowboat
   ./start.sh
   

3. Access the app at http://localhost:3000.

To add tools, RAG, more LLMs, and triggers checkout the Advanced section below.

Demos

Meeting-prep assistant

Chat with the copilot to build a meeting-prep workflow, then add a calendar invite as a trigger. Watch the full demo here.

Customer support assistant

Chat with the copilot to build a customer support assistant, then connect your MCP server, and data for RAG. Watch the full demo here.

Personal assistant

Chat with the copilot to build a personal assistant. Watch the full demo here.

Advanced

1. Native RAG Support: Enable file uploads and URL scraping with Rowboat's built-in RAG capabilities – see RAG Guide.

2. Custom LLM Providers: Use any LLM provider, including aggregators like OpenRouter and LiteLLM - see Using more LLM providers.

3. Tools & Triggers: Add tools and event triggers (e.g., Gmail, Slack) for automation – see Tools & Triggers.

4. API & SDK: Integrate Rowboat agents directly into your app – see API & SDK docs.

Refer to Docs to learn how to start building agents with Rowboat.