wechat-ai-assistant-public
Enables reading target conversations and visible messages on macOS through Accessibility and Vision OCR, with support for controlled reply submission and read-back verification.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@wechat-ai-assistant-publicShow recent messages from my chat with Alice and summarize them"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
微信 AI 聊天助手
这是一个面向 macOS 微信桌面端的本地优先 AI 助手参考实现,展示从会话读取、上下文处理、模型编排到受控回复提交的完整工程链路。
本仓库由真实项目的可公开代码整理而成。联系人、地点、账号、本机路径、运行记录和个性化消息样本均已替换为通用示例;原始私有仓库及本地运行数据不会公开。
核心能力
通过 macOS Accessibility 与 Vision OCR 识别目标会话和可见消息。
提供只读历史、当前会话和受控发送三类 MCP 入口。
使用候选锁定、草稿核验、单次提交和发送读回保障回复幂等性。
对会话记忆进行加密存储、分页检索、敏感信息过滤和冲突处理。
支持
dry-run、observe、supervised-send、live分阶段运行模式。使用 TypeScript、Zod、Vitest 与 Swift 完成类型约束、业务测试和原生桥接验证。
Related MCP server: WeChat FastBridge
安全边界
默认模式为
observe,只读且不发送消息。真实发送必须经过明确授权、草稿验证与发送结果读回。
密钥、会话数据和个人资料不写入仓库;本地运行目录应保持
0700权限。不提供自动登录、盲坐标点击、验证码处理或绕过系统权限的能力。
本地验证
npm ci
npm run lint
npm run typecheck
npm test
npm run native:test项目要求 Node.js 20,并在 macOS 上使用 Swift 工具链验证原生桥接部分。npm test 使用串行文件模式,避免发布事务类测试与清理 dist/ 的测试并行运行时互相影响。
目录
src/:会话、记忆、策略、MCP 与运行时核心逻辑。native/:macOS OCR、窗口识别和受控输入桥接。tests/:单元测试、集成测试和边界场景验证。scripts/:构建、安装和受控发布脚本。config/:脱敏后的发布权限边界示例配置。assets/:脱敏后的测试图片素材。prompts/:不含个人信息的通用提示模板。
脱敏说明
公开版使用“示例用户”“示例联系人”“示例城市”等占位数据。仓库不包含原项目的私人联系人、真实地点、聊天记录、运行日志、验证回执、项目记忆或本机凭据。
许可
本仓库暂未添加开源许可证,仅作为作品集和技术交流展示;未经作者许可,请勿直接复制、商用或用于真实账号自动化。
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP connector for iMessage & Contacts via a local Mac agent + Vercel relay
Let Claude or ChatGPT search, read and send your WhatsApp messages over MCP. OAuth sign-in.
Your own WhatsApp as an MCP server: read, search and send from any MCP client.
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables automation of WeChat on macOS through the Accessibility API, allowing LLMs to fetch recent messages from contacts and send replies based on conversation history.52 PyPI259MIT
- AlicenseCqualityBmaintenanceLocal macOS MCP server for verified WeChat reading, sending, media, and token-efficient allowlisted monitoring. Its Docker image supports registry introspection only; real WeChat automation requires macOS Accessibility.67MIT
- AlicenseNot gradedqualityDmaintenanceA production-oriented MCP server that exposes task-level WeChat Desktop automation to AI clients, enabling status checks, message fetching/reply, contact addition, and Moment publishing on macOS via accessibility automation.4MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP server that lets an LLM read and drive the macOS WeChat client through the Accessibility API, without reverse-engineering or injected code. It supports listing chats, opening conversations, reading messages, sending actions, and saving media locally.MIT