wechat-colleague

Decrypt WeChat 4.0 databases into readable chat data and AI colleague skills.

62|13|Updated Apr 14, 2026
One-click install
npx skills add https://github.com/zhx8702/wechat-colleague --skill wechat-colleague
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wechat-colleague
Source: https://github.com/zhx8702/wechat-colleague/tree/main
Command: npx skills add https://github.com/zhx8702/wechat-colleague --skill wechat-colleague

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pycryptodome, zstandard, flask, pymem, psutil, sqlite3.

What problem does it solve?

This Skill solves the problem of turning encrypted WeChat 4.0 local chat databases into readable conversations and usable AI coworker/character prompts, without you manually writing decryption or extraction code.

Core Features & Use Cases

  • Decrypts WeChat 4.0 local encrypted databases: extracts per-database keys from the running Weixin.exe process (Windows) and uses documented AES-256-CBC page decryption to produce queryable SQLite databases.
  • Views and searches chat records via a Web viewer: generates a Flask-based Web UI that lists sessions, renders messages (including zstd-compressed content), parses quote/reply XML, and supports search.
  • Distills a target person into an AI Skill (Work + Persona): analyzes the target’s messages using provided analyzer templates, then generates colleague-specific SKILL.md plus work.md/persona.md and knowledge/messages content for direct invocation.

Quick Start

Tell the AI: “Run /wechat-colleague setup to detect your WeChat data directory, extract decryption keys, and decrypt the databases so I can sync and view messages.”

Frequently Asked Questions about wechat-colleague

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I decrypt WeChat 4.0 local database files on Windows?

To decrypt WeChat 4.0 local databases, you must extract per-database AES-256-CBC keys from the running Weixin.exe process memory. The Skill automates this extraction and applies documented page decryption rules to produce queryable SQLite databases.

Can I view and search decrypted WeChat chat logs through a Web UI?

Yes, you can view decrypted WeChat chat logs via a generated Flask-based Web viewer. It lists sessions, renders messages including zstd-compressed content, parses quote and reply XML, and supports full-text search.

How to distill a specific person's WeChat messages into an AI persona?

You distill a target person's WeChat messages into an AI persona by analyzing their chat history with provided analyzer templates. The Skill then generates a colleague-specific SKILL.md alongside work.md, persona.md, and knowledge content for direct AI invocation.

Does WeChat chat decryption work on macOS and WSL?

WeChat chat decryption and per-person distillation support Windows, macOS, and WSL environments. The Skill uses a YAML-defined workflow that reads documented encryption and schema rules to generate runtime scripts across these platforms.

What's the best way to sync new WeChat messages incrementally?

The best way to sync new WeChat messages incrementally is by running the Skill's sync command after the initial setup. It processes the decrypted SQLite databases to append new chat records to your existing readable dataset.

Why does the Skill need pymem and pycryptodome to read WeChat databases?

The Skill needs pymem to read the live Weixin.exe process memory for extraction keys, and pycryptodome to perform the AES-256-CBC page decryption required to unlock the encrypted local SQLite databases.