jev-xianhui-windows
Windows version of XianHui that OCRs the visible WeChat window locally and uses Jev to rank which messages must be answered first.
Overview
XianHui for Windows captures the WeChat window, reads the other party's messages with local RapidOCR, asks Jev which ones must be answered right away, and shows results in a floating edge window sorted into reply now (red), soon (yellow) and later (grey), up to five at a time. Clicking an item switches to the matching chat app. It uses the same questions and tiering rules as the Android version, at about one second and $0.00004 per judgment according to the README.
It only judges and never writes replies. Screenshots stay in memory and are never written to disk or logs, message text is sent to the model only at judgment time, there is no telemetry, and the only network call is a single POST. It does not inject into, hook or decrypt WeChat and does not send messages. Known limits: the WeChat window must stay visible because Windows has no system notification API for it, WeChat layout changes may break parsing, images cannot be read, and the list is cleared on restart by design. An installer is available, or it can run from source with Python 3.10+. It is MIT licensed and documented in Chinese.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Score | How urgent is this message (1-9)? | — |
| Yes / no | Does this message need an immediate reply? | — |
How it works
- Capture visible WeChat window into memory
- Local RapidOCR extracts the other party's messages
- Jev judges priority with the shared question set
- Floating window shows red, yellow or grey tiers; click to jump
Strengths
- Local OCR; screenshots never stored
- No telemetry; single network call
- Auto-start option without admin rights
Caveats
- WeChat window must be visible
- Layout changes may break parsing
- Cannot read images or stickers
Getting started
Install XianHui-Setup.exe, enter an OpenRouter key, keep the WeChat chat visible and click Start; or run `python main.py` from source.At a glance
- Type
- Project
- Status
- beta
- Language
- Python
- Runtime
- Windows desktop app
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
先回｜消息优先级助手 Windows + Android双端工具 JEV规则引擎 + AI自动判定消息优先级.
hellogumbo/awesome-jev
Frequently asked questions
Why must WeChat stay visible?
Windows has no system notification interface for WeChat, so it must capture the rendered window.
Are screenshots saved?
No, they stay in memory and are never written to disk or logs.
How does it relate to the Android version?
Same questions and tiering rules, different message capture path.
Related projects
Android and Windows message-priority assistant where Jev scores urgency and reply-now for each message and sorts them by colour.
ProjectbetaKotlinLocal-first work-message classifier that uses Jev to sort chat messages into urgent, to-do, worth reading and skip, with cross-chat to-dos.
ProjectbetaJavaScriptSelf-hosted Feishu bot and web workbench for Jev choices with probabilities, plus experimental word-by-word writing.
ServiceprototypeTypeScriptOpen-source AI email assistant; listed as using Jev as an optional classifier for email intent and triage.
ProjectproductionTypeScriptChrome extension that groups and orders all tabs in a window by meaning and your rules with a single Jev request.
ServiceproductionTypeScriptLarge Claude Code plugin of skills, agents and hooks that can optionally ask Jev to classify a session's work category.
ProjectproductionTypeScript