🤖 Agent1 - Your Local AI Assistant

A local-first personal AI agent that runs on your machine. No cloud, no account required.

✨ What It Does

• Runs AI agents locally - Your data stays on your device
• Connects to Ollama - Uses local AI models (Llama, Mistral, etc.)
• MCP support - Works with 80+ tools (filesystem, Git, Slack, etc.)
• Desktop app - Clean UI for managing agents
• Persistent memory - Remembers conversations

🚀 Quick Start

Start the Server

cargo run --bin agent1 -- server --bind 127.0.0.1:17371

Open the Desktop App

cd desktop
npm install
npm run tauri:dev

💻 CLI Commands

Command	What It Does
agent1 run --agent my-agent.toml --task "hello"	Run an agent
agent1 server	Start the API server
agent1 models	See available AI models
agent1 memory write "note"	Save a note

🌐 API Endpoints

Endpoint	Description
GET /api/agents	List your agents
POST /api/sessions/run	Run a task
GET /api/sessions/{id}/stream	Stream progress
GET /ws/events	Real-time events

⚙️ Configuration

Connect to Ollama

export OLLAMA_BASE_URL=http://localhost:11434

Use OpenAI-Compatible APIs

export OPENAI_BASE_URL=http://localhost:8000/v1

🔧 Troubleshooting

Can't connect to Ollama?

Run ollama serve in another terminal.

Port in use?

Try: cargo run --bin agent1 -- server --bind 127.0.0.1:17372

MCP server not working?

Check: agent1 mcp list

🔒 Privacy

• Runs locally - data never leaves your machine
• API binds to localhost only
• Secrets are automatically redacted in logs