wechat_mcp_server
MCP server from Lisheri/wechat_mcp_server
claude mcp add --transport stdio lisheri-wechat_mcp_server docker run -i image-name \ --env DESCRIPTION="Go + Python MCP server for WeChat Mini Program data crawling (placeholder if using a container)."
How to use
This MCP server provides an automated data collection workflow for Mac-focused WeChat Mini Programs. It leverages a Go-based backend to orchestrate the crawling pipeline and a modular Python crawler to perform window management, UI interaction, screenshot capture, image analysis, and data aggregation. When running, the server coordinates between modules to detect the WeChat window, navigate to a mini program, capture targeted screenshots, run OCR and image analysis, and output a structured JSON report that summarizes functionality, navigation paths, and detected features. Users can inspect logs and the generated crawl results to understand how the miniature app behaves and what data was extracted.
How to install
Prerequisites:\n- macOS (optimized) with remote automation permissions enabled\n- Go 1.19+ (for the backend server)\n- Python 3.7+ (for the crawler modules)\n- Git\n\n1) Clone the repository:\n\nbash\ngit clone <repository-url>\ncd mcp_wechat_mini\n\n\n2) Install Go dependencies (server):\n\nbash\ngo mod tidy\n\n\n3) Install Python dependencies (crawler):\n\nbash\ncd py_scripts\n./install_python_dependencies.sh\n# or install manually using pip if script is unavailable\npip3 install -r requirements.txt\n\n\n4) Start the MCP server (Go backend):\n\nbash\ngo run main.go\n,\n\n5) Run the crawler (Python):\n\nbash\n# New modular crawler flow\npython3 run_crawler.py\n# Or run the modular Python script directly\ncd py_scripts && python3 main.py\n
Additional notes
Tips and common considerations:\n- Ensure macOS automation permissions are granted for Python to control the screen and mouse\n- The crawler relies on precise window positioning; adjust MINI_PROGRAM_BOUNDS and entry points as needed in py_scripts/config.py\n- If you encounter window detection issues, verify that the WeChat window is visible and not obscured by other applications\n- For best results, run on a stable macOS environment with consistent screen resolution.\n- Logs and JSON outputs are typically written to crawl_results and log files; check the configured paths in the server and crawler configs.\n- When using a container, replace image-name with an actual image that bundles both the Go server and Python environment, and ensure proper volume mounts for logs and outputs.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP