jev-chat: Multi-Platform Chat Decision Assistant and Intent Analyzer

jev-chat: Multi-Platform Chat Decision Assistant and Intent Analyzer

GitHub Open Source
jev-chat/jev-chat-jarvis

Read chats, judge intent, and draft replies using your own API keys across Android, Windows, macOS, and iOS.

Repository snapshot

  • Repository: jev-chat/jev-chat-jarvis
  • Language: Kotlin
  • License: MIT
  • Stars: 8 373
  • Forks: 1 335

Daily open-source picks: GitHub Open Source on Telegram

What it is

jev-chat is an open-source chat decision assistant designed to read conversations, judge intent and risk, and draft context-aware replies. Available across Android, Windows, macOS, and iOS, the project aims to help users navigate complex or high-volume messaging without automating away human final control. Rather than blindly generating text, it prioritizes a judgment phase to understand what the other person actually means before suggesting how to respond.

Why it matters

When a message lands, the difficult part is rarely typing the response; it is accurately gauging the tone, subtext, and potential friction of the interaction. Traditional AI integrations jump straight into drafting text, often missing nuance or escalating awkward situations. jev-chat separates judgment from drafting, ensuring that every suggested reply is checked against an explicit assessment of the conversation's risk and intent.

Key features

  • Evaluate message intent and risk levels before drafting any responses.
  • Rank and check generated replies against initial conversation judgments.
  • Integrate across Android, Windows, macOS, and iOS with native overlays.
  • Send all API requests directly to your own configured model provider.
  • Operate via accessibility services or local desktop OCR without server logging.

Quick start

To get started with the source code, clone the repository:

git clone https://github.com/jev-chat/jev-chat-jarvis.git

Who should try it

This project fits developers and power users who handle extensive cross-platform messaging and want to experiment with client-side AI assistants. Because it relies on your own API keys and local device capabilities, it appeals to developers who prioritize data privacy and want to avoid third-party server logging.

Verdict

jev-chat provides a thoughtful approach to chat assistance by putting judgment before generation. However, users should carefully consider platform terms of service when running accessibility overlays or custom keyboards on third-party messaging apps.

Continue exploring

Get the next pick in Telegram: Follow GitHub Open Source

Curated from the repository README and metadata. No hands-on test is claimed; evaluate it in a sandbox before production use.

Report Page