Small decisions. Interesting possibilities.Submit contentSubmit

How it uses Jev

The user long-presses a text message and selects Intent. The plugin sends message and per-conversation relationship context for category choices and urgency/risk yes-no or score judgments, then displays results without sending a reply.

What Jev decides

Analyze one selected WeChat messageExample answers · not a recorded Jev response · Source ↗
Question 1 · intent
YOUR APP
INSTRUCTION

Classify the selected message intent with relationship, gender, and recent-dialogue context.

STATE

One pending text message plus the conversation relationship, counterpart gender, and recent conversation context. Selections shown are illustrative, not observed outputs.

JEV · CHOICE
  1. 打招呼或闲聊
  2. 询问信息或情况
  3. 请求帮忙或让我办事
  4. 催促进度或催回复
  5. 商务合作或推销
  6. 投诉或表达不满
  7. 通知或告知信息
  8. 表达情绪或吐槽
  9. 邀约见面或请客
  10. 其他
Question 2 · urgency
YOUR APP
INSTRUCTION

Rate urgency using the source-defined 0 to 3 scale.

STATE

One pending text message plus the conversation relationship, counterpart gender, and recent conversation context. Selections shown are illustrative, not observed outputs.

JEV · SCORE
03
Question 3 · need reply
YOUR APP
INSTRUCTION

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

STATE

One pending text message plus the conversation relationship, counterpart gender, and recent conversation context. Selections shown are illustrative, not observed outputs.

JEV · NOUL
YesNo
Question 4 · risk
YOUR APP
INSTRUCTION

Assess whether the message presents actual risk to the user, considering the relationship.

STATE

One pending text message plus the conversation relationship, counterpart gender, and recent conversation context. Selections shown are illustrative, not observed outputs.

JEV · NOUL
YesNo
Question 5 · advice
YOUR APP
INSTRUCTION

Choose the most suitable response action.

STATE

One pending text message plus the conversation relationship, counterpart gender, and recent conversation context. Selections shown are illustrative, not observed outputs.

JEV · CHOICE
  1. 立刻回复并给出明确答复
  2. 简短确认收到,稍后详答
  3. 先问清楚细节再答复
  4. 礼貌婉拒
  5. 不必回复
  6. 涉及钱/账号,先人工核实再回
  7. 需要上报或转给相关同事
  8. 共情或闲聊式回应即可
Question 6 · emotion
YOUR APP
INSTRUCTION

Choose the closest emotion in relationship and recent-dialogue context.

STATE

One pending text message plus the conversation relationship, counterpart gender, and recent conversation context. Selections shown are illustrative, not observed outputs.

JEV · CHOICE
  1. 平静
  2. 高兴
  3. 不满
  4. 生气
  5. 焦虑
  6. 着急
  7. 难过
  8. 讽刺
  9. 客套
Question 7 · reply style
YOUR APP
INSTRUCTION

Choose the suitable reply posture, without deciding wording.

STATE

One pending text message plus the conversation relationship, counterpart gender, and recent conversation context. Selections shown are illustrative, not observed outputs.

JEV · CHOICE
  1. 建议调侃打趣
  2. 建议正常交流
  3. 建议礼貌客套
  4. 建议肯定赞赏
  5. 建议关心问候
  6. 建议共情吐槽
  7. 建议表达同情
  8. 建议先安抚道歉
  9. 建议先问清细节
  10. 建议温和婉拒
  11. 建议暂不回复

App workflow

  1. Select message and relationship

    Long-press a text message; optionally cycle the relationship setting for that conversation.

  2. Run analysis

    Choose Intent from the menu. The plugin sends its typed analysis request using the configured Jev key.

  3. Read the result

    The plugin displays intent, emotion, urgency, and suggested posture in fixed toast output; it does not compose or send a reply.

Why it is interesting

A lightweight UI gesture turns relationship-sensitive message interpretation into compact, typed judgments while keeping the final response with the user.