WeChat Butler v0.1.0

AI Service Layer for the WeChat ecosystem. Connects to chatshell-api via MCP for data queries, provides LLM-powered AI chat with tool_call loop, and streams results to chatlog-session via SSE.

Quick Start

# Create conda environment
conda env create -f environment.yml

# Activate
conda activate wechat-butler

# Configure
cp .env.example .env
# Edit .env with your API keys

# Run
python -m wechat_butler

Configuration

See config.yaml for all options. API keys are set via environment variables in .env.

API

• POST /api/v1/ai/chat - AI chat (SSE streaming)
• GET /api/v1/ai/models - List models
• GET /api/v1/ai/status - Service status
• GET /api/v1/ai/config - Get config (masked keys)
• POST /api/v1/ai/config - Update config
• CRUD /api/v1/ai/prompts - Prompt management
• GET /health - Health check (no auth)

All endpoints except /health require X-Butler-API-Key header.