User repository on GitHub
Jev辅助判断的对话聊天助手
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Tools & CLIs
Jev 对话助手 (Jev Chat Assistant)
A Windows desktop assistant that OCRs a selected chat area, lets you correct the transcript, then has Jev analyze the conversation and rank three candidate replies from an optional reply model; nothing is auto-sent.
- Stars
- 8
- Language
- Python
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- qcjlchina
- Last verified
- Oct 2026
Highlights
- v1.4.0 adds proofread-before-analysis, reply purposes such as polite refusal and de-escalation, and per-candidate rewrites that clear old scores.
- Analyzes all messages by default or the last 10/20/50; selected text and background are capped at 20,000 characters with no silent truncation.
- OCR runs locally; chat text goes to TypeSafe Jev only when you confirm analysis, and API keys live in Windows Credential Manager.
- A safety check blocks transaction content such as transfers, red packets, receipts and payments; candidates are view-and-copy only.
- Built on Liyucheng1997/332_lab-jev-chat; install.ps1 and start.ps1 target Windows 10/11 x64 with Edge WebView2.
Quickstart
bash
powershell -ExecutionPolicy Bypass -File .\windows\install.ps1
powershell -ExecutionPolicy Bypass -File .\windows\start.ps1Watch out
Copy reviewed against the repo README on 2026-10-15; the README is in Simplified Chinese with translations linked, and descriptive claims such as the transaction safety check are not independently tested here. Images, voice messages and quoted cards cannot currently be reliably restored.
Questions about this entry
Short answers drawn from the linked project or page.
A Windows 10/11 desktop app, at v1.4.0, that selects a chat region on screen, OCRs it locally, lets you correct the transcript, and then asks Jev to analyze the conversation and rank three candidate replies produced by an optional reply model.
Jev performs the conversation analysis, and when a reply model is configured it also ranks the three generated candidates. Reply purposes include general reply, polite refusal, clarifying a misunderstanding, ending a topic and de-escalating a conflict.
No. The README states the program does not read chat databases, operate chat input boxes or auto-send messages: candidates are for viewing and copying only. OCR runs on your machine, and the recognized text is sent to TypeSafe Jev when you confirm the analysis.
Windows 10/11 (x64) with the Microsoft Edge WebView2 Runtime, plus Python 3.12, Node.js 20+ and npm if you run or build from source. First use needs a TypeSafe/Jev API key; a reply-model API key is optional.
From the repository root in PowerShell, run install.ps1 and then start.ps1. A prebuilt v1.4.0 Windows release is also linked from the README, and source builds produce an executable under dist that needs its _internal folder kept alongside it.
A transaction-content safety check intercepts transfers, red packets, receipts and payments across the full input, including unselected messages. Images, voice messages and quoted cards cannot currently be reliably restored, and analysis covers at most 20,000 selected characters.
OCR happens locally; recognized text goes to TypeSafe Jev on analysis and to the reply model if enabled; update checks only query GitHub for the latest version number. API keys are stored in Windows Credential Manager, while settings.json holds only non-secret settings.
More Jev resources like this
Traffic signal control with Jev applied to SUMO simulation: each control step packs the road network into one state with a choice question per intersection, and the model's chosen phase, distribution and confidence drive the signal with local-ranking fallbacks.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
usail-hkust/jevlightPython
usail-hkust/jevlightPythonToolkit, knowledge base and local dashboard for building with Jev via OpenRouter Alpha Decisions or the TypeSafe Direct API: a zero-dependency SQLite request log, a localhost REST server and Python CLI demos.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
rileycarney/jevtoolsHTML
rileycarney/jevtoolsHTMLA Windows fork of the jev-chat WeChat assistant that fixes DPI-unaware screen capture on scaled displays and adds an optional key-free local laya judge — with the author's own measurements showing that local engine is not yet usable for Chinese chat.
Jev Tools & CLIs#community#catalog-import#logicrw
A TypeScript decision engine and visual studio that wraps TypeSafe Jev: developer-authored loss matrices price mistakes, exact Total Variation distance stress-tests probabilities, and clarifying questions are ranked by Value of Information minus interruption cost.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
satiricalguru/hinge-jevTypeScript
satiricalguru/hinge-jevTypeScriptLocal, inspectable image perception for the text-only Jev: OCR text in reading order plus the same words on a character grid become the state Jev decides over, returned as a plain dict. CLI, MCP server and agent skill.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
leddoengano/jev-eyesPython
leddoengano/jev-eyesPythonModeration for communities and apps, powered by Jev: every message gets a probability for spam, scam, harassment, nsfw, doxxing and more, plus rules in plain English. Thresholds and actions are yours, and flag-only is the default.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
ohernandezdev/jevmodPython
ohernandezdev/jevmodPythonFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.