AI Agent Skill-WeChat AI Message Processing System

WeChat AI Message Processing System

Build a Local AI Pipeline for Group Chat Analysis, Investment Radar & Contact Intelligence

Turn your WeChat chat history into an AI-accessible knowledge base — completely offline, fully local, privacy-first.

Overview

WeChat stores years of valuable conversations locally — but that data is encrypted and inaccessible. This skill provides a complete, step-by-step blueprint to unlock it, pipe it into AI tools like CherryStudio, and run custom agents powered by DeepSeek, Gemini, or any LLM of your choice.

The Pipeline

WeChat Client → chatlog Decryption → HTTP + MCP (:5030) → CherryStudio → DeepSeek/Gemini

What You'll Build

Group Chat Summarizer: Auto-generate daily/weekly digests of any group conversation.

Investment Radar: Scan stock-trading groups for ticker mentions, sentiment shifts, and consensus signals.

Contact Intelligence: Map conversation frequency, topic clusters, and relationship networks.

Custom AI Agents: Design purpose-built agents with tailored prompts for your specific use cases.

What's Included

SKILL.md — Complete step-by-step build guide with all commands and configurations.

Tool comparison — Detailed evaluation of chatlog forks (myysophia/wechat-log vs chatlog_alpha vs chatlogwebUI).

China network fixes — Go proxy configuration and compilation workarounds for mainland users.

MCP setup guide — Connect WeChat data to CherryStudio, Claude Code, or any MCP-compatible client.

Requirements

• Windows PC with WeChat installed

• Go 1.26+, GCC (MinGW), Python 3.10+, Node.js 18+

• VS Code with GitHub Copilot Chat

Note: This skill is technical — you'll be compiling Go code and configuring local servers. But the AI guides you through every command. All processing is local; your chat data never leaves your machine.