Small decisions. Interesting possibilities.Submit contentSubmit

How it uses Jev

JevGuide reads a WeChat conversation through Android accessibility or a screenshot reader, combines it with contact context and optional local memory, and sends a string state to Jev. Typed questions judge intent, urgency, reply expectation, risk and relationship progress. The app displays judgments; a separate chat model can draft three replies.

What Jev decides

Source-defined request (illustrative)Example answers · not a recorded Jev response · Source ↗
Question 1 · need reply
YOUR APP
INSTRUCTION

Is the other person currently waiting for a reply, considering the relationship and conversation context?

STATE

A string containing the extracted conversation, contact relationship context and optional local-memory digest. The source preserves original Chinese conversation text even when question instructions are English.

JEV · NOUL
YesNo

App workflow

  1. Read the conversation

    A manual or configured automatic trigger reads accessibility text, falling back to screenshot transcription when needed.

  2. Judge the context

    The analyzer combines contact data and memory into one string state and asks typed Jev questions.

  3. Show advice

    The app shows judgments and progress, optionally requests draft replies from a chat model, and stores the local analysis.

Why it is interesting

The app separates conversational judgment from wording. Its persistent progress overlay uses a typed score and local contact history, while the user retains control of sending messages.