zjp1997720 / wechat-radar

What is wechat-radar?

WeChat Radar is a local-first intelligence dashboard that turns noisy WeChat group chats into actionable briefings. It runs on your Mac, reads your actual WeChat messages through a companion CLI tool, and stores everything in a local SQLite database that never leaves your machine. The interface shows you daily summaries, cross-group topic clusters, shared links and tools, @mentions, and per-group activity reports. Built with Next.js and TypeScript, it requires macOS with WeChat 4.x and a running daemon process.

Why is it gaining traction?

The hook is simple: if you live in WeChat groups, you know the signal-to-noise ratio is brutal. This tool gives you a daily digest of what actually matters across dozens of groups without opening any of them. The topic clustering feature uses Codex CLI to automatically surface cross-group themes. Link intelligence deduplicates shared articles and tools with generated titles, so you can see what resources are actually circulating. The copy-friendly daily reports are designed to paste directly into AI tools for further processing. Everything stays local by default, which addresses the privacy concerns that would otherwise make this a non-starter.

Who should use this?

This is for developers and power users who manage or participate in many WeChat groups and need to stay on top of information without drowning. Community managers tracking discussions across multiple groups, technical leads monitoring AI/developer communities, or anyone who gets value from cross-referencing what is being discussed across different circles. If you already use wx-cli and want a visual layer on top of your chat data, this fills that gap. Not for casual users or anyone unwilling to run a local daemon alongside WeChat.

Verdict

WeChat Radar solves a real problem for a specific audience, and the local-first architecture is the right call for this use case. The feature set is thoughtful and the demo mode makes it easy to evaluate. However, with only 49 stars and a credibility score of 0.8%, this is early-stage software from a solo maintainer. The documentation is decent but test coverage is unclear, and the dependency on wx-cli plus macOS-only support limits the audience. Worth trying if you fit the profile, but do not bet production workflows on it yet.