User repository on GitHub
微信里的话不知道怎么接？macOS 悬浮聊天助手：识别消息意图与沟通风险，GPT 生成多种话术，Jev 评估候选，一键填入微信。话我帮你想，发送你来定。
- Forks
- 6
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Tools & CLIs
Jev Yaba Wechat
macOS floating WeChat companion that reads the visible chat with on-device Vision OCR, judges intent and reply risk with Jev, drafts toned replies with GPT, and only fills text after you approve it.
- Stars
- 24
- Language
- Python
- License
- MIT
- Published by
- Community
- Author
- wuxie888
- Last verified
- Oct 2026
- Languages
- Python
Highlights
- Floating panel tracks the WeChat window with collapsible, pausable on-device screen reading.
- Judges eight intent classes on recent context and shows risk plus action guidance before replies.
- Up to ten tones with three active groups and six ranked candidates per incoming message.
- Copy or accessibility-based fill with read-back check; never injects into WeChat or sends alone.
Watch out
Source-only release for macOS 13 and above with no installer; local judgment needs a large first-time model download, OCR can misread speakers and cards, and image or voice meaning is unreliable. 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. Selecting a candidate only copies it or fills the WeChat input box through accessibility features. Sending always happens in WeChat by you, and missing input controls fall back to copy plus paste.
Jev handles judgment: intent among eight classes, a 0-9 risk score, and candidate ranking. GPT handles wording across selectable tones. Without GPT there is judgment but no drafted replies.
No. It is macOS 13 and above from source, started with a launch script, and there is no downloadable installer. Authorization targets the terminal or Python when run from source.
Window pixels are OCRed locally; configured GPT and TypeSafe calls send message text and tone prompts to those services. Keys live in a local config file outside the repo and inputs mask entry.
Open WeChat and enter the chat you want, then run ./start.command. The first start prepares Python and its dependencies and needs network; grant Screen Recording to read the WeChat window and restart afterwards, and grant Accessibility only if you want the Fill action, since copying needs neither.
Open Settings from the panel or menu bar, fill the GPT base URL, model name and API key, choose Responses or Chat Completions, and Test GPT; then fill the Jev base URL, model and key and Test Jev. The two tests are separate, passing one does not mean connected, and after both pass you pick 正常回复 and Save and restart.
Ten built-in tones include 高情商话术, 贴吧老哥 v1.0, 拒绝加班, 卑微乙方 and 稳如老狗; up to three are active at once and each request generates two replies, one steady and one more individual, for up to six candidates. Add custom ones with JEV_TONES as name=description pairs separated by |, and a same-name setting overrides a built-in tone.
More Jev resources like this
39GitHub 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/jevclipPython547GitHub stars
Live SEO audit that crawls a site from one homepage URL, checks 52 rules tied to Google Search Central, and asks Jev typed questions about every page to rank fixes.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
AgriciDaniel/jev-seoPython
AgriciDaniel/jev-seoPython30GitHub stars
Rust CLI that checks each PDF page for a usable text layer, extracts text pages locally, sends only the rest to an OCR model via LiteLLM, and merges a page-ordered result.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
misbahsy/doc-routerRust
misbahsy/doc-routerRust27GitHub stars
Local-first trading journal and review harness that is read-only over markets, captures runs into frozen envelopes and evidence packs, and requires explicit human promotion before any rule returns.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
myc0576/SmartMoney-CubPython
myc0576/SmartMoney-CubPython3GitHub stars
Recruiter dashboard that applies calendar date math plus visa, tenure, license and clearance gates, then a Jev decision step, sorting resumes into disqualified, advance or manual review.
Jev Tools & CLIs#community#jevbest#catalog-import
5GitHub stars
A point-in-time backtest of a William O'Neil momentum strategy where jev picks the entries and decides the exits, paired against the same rules with no model — and honest enough to report the model as indistinguishable from them.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
michaelpersonal/jev-trade-ccPython
michaelpersonal/jev-trade-ccPythonFrom 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.