pi-jev-reply
Pi extension where Jev decides whether a reply needs rewriting, then your model clarifies it and may add a small visual.
Overview
Pi Jev Reply is a finishing pass for replies in the Pi terminal. With a Jev key, Jev first decides whether an edit would help; if so, your current or selected model rewrites the reply in plain language and may add a compact table, diagram or text chart. If the edit fails or is invalid, the original reply stays. Without a Jev key, enabled rewriting goes straight to the model and review thresholds do not apply.
Visuals are added only for recognised explicit requests when enabled, and a request to omit visuals takes priority. Structural checks reject edits that drop protected references or change numbers, though they do not guarantee factual accuracy. A welcome page and local settings service open on first interactive launch; settings cover model, rewrites, visuals, instructions and language (English default, Chinese available). It requires Pi 0.85.1+ and Node 22.18+.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Would an edit make this reply clearer? | yes · no |
How it works
- Agent produces a reply.
- Jev decides whether a rewrite would help.
- Model rewrites and optionally adds a visual.
- Structural checks accept the edit or keep the original.
Strengths
- Original reply kept on any failure.
- Numbers and references protected by checks.
- Works without Jev too.
Caveats
- Rewrites can still introduce factual errors.
- Adds model calls for rewritten replies.
Getting started
pi install npm:@each1024/pi-jev-reply, then /reload.At a glance
- Type
- Plugin
- Status
- beta
- Language
- TypeScript
- Runtime
- Pi extension
- Has code
- yes
Use cases and integrations
How curated lists describe it
Pi extension: clearer replies via Jev review + optional rewrite/visuals.
hellogumbo/awesome-jev
Frequently asked questions
What does Jev decide?
Whether a reply would benefit from a rewrite.
What if the rewrite fails?
The original reply is kept.
Do I need a Jev key?
No, without one rewriting goes directly to the model.
Related projects
macOS WeChat helper where an LLM drafts replies and Jev classifies intent, rates risk and ranks the candidates; you press send.
ProjectprototypeSwift- jev-chat-windows★ 802
Windows WeChat reply assistant: local offline OCR reads the chat, Jev judges intent and tension and ranks three drafted replies to fill in.
ProjectproductionPython AstrBot plugin where Jev decides whether the bot should join a group conversation; the main model still writes the reply.
ProjectproductionPythonDeepSeek Harness plugin that gives the human a Jev second opinion beside the agent's multiple-choice questions.
ProjectbetaTypeScriptLocal-first Windows writing studio with knowledge graphs and author voice, where optional Jev evaluates candidates and editorial decisions.
ProjectbetaTypeScriptiOS custom keyboard that shows a copied message's intent, risk level and candidate replies inside any chat app, filling but never sending.
AppbetaSwift