agentleFS
ITch8ITch8/customs-data-skills/skills/conversation-context/SKILL.md

conversation-context

Skill9 starsChanged 24 days agoMITView on GitHub

跨 Email 与 WhatsApp 的统一客户沟通上下文与 Trade Case 解析。用户要梳理客户沟通历史、把邮件与消息归入同一 Trade Case 时使用。

The file

Files
SKILL.md

Conversation Context

核心对象

Conversation 是 Customer 与企业围绕一个业务主题的连续上下文;Message 是不可变的渠道消息;Email 与 WhatsApp 只是 channel,不是 Customer 或 Trade Case。

Context

分析时读取 Customer、Contact、Trade Case、Conversation、最近消息、TradeActivity、Knowledge/Evidence、Draft、Quote、Open Questions 与 Next Action。

Case Assignment

已有 Conversation、唯一活跃 Case 或强显式产品引用可作为 high confidence;多个可能 Case 为 medium 并要求人工确认;无法可靠判断进入 unassigned。禁止只凭公司名自动合并客户。

Requirements / Conflicts

跨渠道需求必须保留来源 message IDs。数量、交期、价格等冲突不能自动覆盖,进入 conflict review。

Human Boundary

禁止自动发送 Email/WhatsApp、自动批准报价、自动修改 Case 关键事实、将 AI inference 当作业务事实。所有外部 Draft 必须通过 Business Response Guard 与 Human Review。

Privacy

消息全文仅用于授权的上下文;普通日志只保存引用 ID、状态、延迟和错误码,不记录客户消息全文。原始消息不可变。

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.