zhuobichen/weflow-cli
Developer toolsThis page was translated automatically. Technical details may differ from the original. View the original
Local-first WeChat data tools: chat record query export, official account daily reports and personal knowledge base (MCP compatible)
Overview
: Read and query chat records locally, export readable HTML; Organize official account articles into daily reports with AI summaries; Read WeChat favorites; Build a knowledge base capable of semantic search; And connect with AI editors such as Claude Code via MCP. Supports Windows and Linux. 🔒 : Runs entirely on your computer—zero telemetry, zero cloud reporting; Database keys are encrypted and stored locally via machine-bound AES-256-GCM. AI features are only enabled after explicitly configuring your own API Key. See the Security and Privacy Statement for details. : Only on devices controlled by you personally will process data you own or have clearly authorized access. Do not use this project to monitor, track, or read others' WeChat data, nor deploy it on others' devices or operate remotely silently. Please confirm applicable laws, WeChat service terms, and account permissions before use; Third-party tools may trigger account risk controls, restrictions, or bans. This project does not constitute legal advice; users bear their own responsibility for use. macOS: Automatic initialization is not currently available; If you have obtained local data access credentials through official or other authorized means, please complete the configuration using weflow-cli init --path and config assistance. Linux's automatic initialization capability depends on WeChat distributions, Python dependencies, and current user permissions; Please run weflow-cli check first and configure according to system security policies. Do not lower system permissions just to troubleshoot. : This tool is limited to accessing data on your own device. Without explicit consent from others, you may not use it to monitor your spouse, employees, or any third party, nor deploy it on another person's device or run it silently remotely. Please also pay attention to WeChat's terms of service, account risk control, and local legal requirements. See SECURITY.md for details. npm packages are handled separately from GitHub releases; If you need the latest fix, please first confirm the npm package version or directly use the GitHub master fork code. You need Node.js 18+, Python 3.10+, and a Windows WeChat account you are logged into. After installation, first check the local environment:
Overview (machine translated)
𝗪𝗲𝗙𝗹𝗼𝘄 𝗖𝗟𝗜
Chinese Simplified · English
People today do not see the moon of ancient times, but this moon once shone upon those of old
Heaven and Earth are the inns for all things; Time is the passing guest of a hundred generations.
Bring chat records, official account reading, and personal knowledge workflows back to your local computer.
WeFlow CLI Turns WeChat into Your Local Second Brain: Read and query chat records locally, export readable HTML; Organize official account articles into daily reports with AI summaries (can be sent via email); Read WeChat favorites; Build a knowledge base capable of semantic search (even voice messages can be searched); And integrate AI editors like Claude Code via MCP. Supports Windows and Linux.
It specifically answers a few questions WeChat can’t answer on its own: who is waiting for my reply? Who did I promise or have I done later? My relationship with whom is fading; how a certain account’s views have changed; what were we talking about on this day of that year?
🔒 > Local First: Runs entirely on your computer—zero telemetry, zero cloud reporting; Database keys are encrypted and stored locally via machine-bound AES-256-GCM. AI features are only enabled after explicitly configuring your own API Key. See the Security and Privacy Statement for details.
This project is limited to processing data you have access to. Please comply with applicable laws, WeChat rules, and others’ privacy boundaries.
** Usage ⚠️ Boundaries and Risks**: Only process data you own or have authorized access on devices you control. Do not use this project to monitor, track, or read others’ WeChat data, nor deploy it on other people’s devices or run it remotely silently. Please confirm applicable laws, WeChat service terms, and account permissions before use; Third-party tools may trigger account risk controls, restrictions, or bans. This project does not constitute legal advice; users bear their own responsibility for use.
Overview of Effects
WeChat Favorites Query (reads local favorite.db, supports type filtering and keyword search):
Complete Workflow, One Command, One Step:
Several WeChat questions that can’t be answered (all read local libraries, no internet connection, no model adjustment):
Platform support
| Capabilities | Windows | macOS | Linux |
|---|---|---|---|
| Automatic Data Directory Discovery | ✅ | ✅ | ✅ |
| Local Data Initialization | ✅ Automatic Discovery/Validation | ❌ Manual Credentials Required | ⚠️ Depends on Distribution and Permissions |
| Chat Record Query (NT / Local Database) | ✅ | ✅ | ✅ |
| WCDB Data Services (Contact Nicknames, etc.) | ✅ | ⚠️ Depend on native libraries | ⚠️ Depend on native libraries |
| WeChat Daily / Knowledge Base / MCP | ✅ | ✅ | ✅ |
macOS: Automatic initialization is not currently available; If you have obtained local data access credentials through official or other authorized means, please follow the weflow-cli init --path and config help to complete the configuration.
Quick Start with Linux (WeChat 4.x Linux Native Version)
This page was translated automatically. Technical details may differ from the original.
Install
This server does not publish a one-line install command.
Open the repository installation guide New to MCP? Start with the mcpskill.org guide →