Aimark-dai/jev-chat-windows-deepseek-jev
User repository on GitHub
DeepSeek + TypeSafe JEV 微信回复助手：Windows 正式版、Apple Silicon macOS 预览版；本地 OCR 与人工可控回复
- Forks
- 5
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Tools & CLIs
Jev Chat Windows DeepSeek Jev
WeChat reply helper that reads the visible chat with local OCR, drafts three DeepSeek candidates, and optionally adds TypeSafe pre-judgment plus review sorting. Fill-in never auto-sends unless Windows countdown is explicitly enabled.
- Stars
- 15
- Language
- Python
- Published by
- Community
- Author
- Aimark-dai
- Last verified
- Oct 2026
Works forJev for Classification
Used inJev for Software & DevTools
- Languages
- Python
Highlights
- Local OCR follows the current WeChat session without touching its database.
- Flow is pre-judgment, DeepSeek drafting, review sorting, and at most one rewrite.
- Judgment card shows risk, intent with confidence, needs, and suggested actions.
- Keys stay out of config and logs; Mac paths and Windows folders are documented.
Watch out
Windows is the working build; the Mac preview lacks full WeChat read and fill acceptance and auto-send stays disabled. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
No. TypeSafe outputs structured judgment and probabilities around risk, intent, needs, and actions; the Chinese reply candidates themselves come from DeepSeek. Closing the TypeSafe path still leaves DeepSeek drafting with basic judgment and sorting.
Windows with desktop WeChat is the supported path: unzip the full folder, run the executable, and add a DeepSeek key plus an optional TypeSafe key. Mac Apple Silicon previews need screen-recording and accessibility grants, are unsigned, and must not be trusted for important sends.
Keep auto-send off and copy or fill manually. Even with Windows countdown enabled, switching sessions, new messages, lost foreground, or a failed review cancels the send, and transfer or red-packet content is blocked from drafting.
Window screenshots plus local OCR only, with no injection, database decryption, or memory reads. Frames stay in memory, keys live in environment variables or keychain rather than config and logs, and update checks query releases without chat content.
TypeSafe Jev pre-judges intent, risk, needs, and suggested action, then DeepSeek writes three candidate replies, then Jev reviews quality and ranks them by choice probability. If all candidates fail review it rewrites once; failing again disables auto-send.
Windows 10 1903+ or Windows 11 with WeChat 4.x runs the stable build; a macOS Apple Silicon preview needs macOS 15+. Both use the desktop WeChat client, and the Mac build has not been validated against real Mac WeChat capture and fill.
The WeChat window must stay non-minimized; OCR can misread names or text; the Windows fill can click the wrong conversation when the layout just changed; and the Mac preview is unsigned and force-disables auto-send.
More Jev resources like this
34GitHub stars
Pure-HTTP automation for the TypeSafe register, magic-link login, onboarding, and API-key creation flow, with pluggable temp-mail backends, concurrent batch runs, and key verification exports.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
2951461586/Jev-Register-ToolPython
2951461586/Jev-Register-ToolPython36GitHub stars
Python toolkit plus 0.8B multimodal model that takes a message, document, or image and returns calibrated choice, yes/no, and score answers in one forward pass for local decision workloads.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
PsiACE/dohnutsPython
PsiACE/dohnutsPython39GitHub stars
Video tool where Jev scores subtitle segments and code assembles a short highlights cut plus a de-watered full cut with timestamped summaries.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
cclank/jevclipPython
cclank/jevclipPython30GitHub stars
Forkable weekday radar that asks Jev one yes-or-no question per interest over each new arXiv announcement and publishes a page plus RSS feed.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
Eliot5566/JEV-Paper-RadarPython
Eliot5566/JEV-Paper-RadarPython38GitHub stars
Async supervisory layer for Hermes Agent that locks Definitions of Done, proves mechanical criteria locally, and calls Jev only for unresolved semantic checks under a token budget.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
keeltrace/hermes-nervePython
keeltrace/hermes-nervePython15GitHub stars
Two-tier decision engine pairing a 22.6M-parameter MiniLM encoder on the Apple Neural Engine with the on-device Apple Foundation Model as fallback for uncertain inputs.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
kerryrm/systemANEPython
kerryrm/systemANEPythonFrom the Jev guides
Original write-ups on what this entry is for.
The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.
Jev Tools & CLIs#gateway#vercel#cloudflare
Read article
Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.
Jev Tools & CLIs#open-models#local#onnx
Read article
Jev labels a record that already exists. Code moves, flags, or holds it. How the document, mail, and transfer tools keep the action out of the model.
Jev Tools & CLIs#classification#email#security
Read article
From the Jev community
Posts from builders shipping with Jev right now.