# Conversation Summary

> 会话小结 - 输入对话内容，调用API返回会话摘要，支持增量更新历史摘要。

- Skill: `kernel8901/conversation-summary` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds@latest add kernel8901/conversation-summary`
- Raw SKILL.md: https://api.skillmd.com/api/skills/kernel8901/conversation-summary/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Integrations & APIs
- License: MIT
- Author: kernel8901 (https://skillmd.com/u/kernel8901)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/kernel8901/conversation-summary

---


# Conversation Summary - Agent Instructions

Use this skill to generate summaries for conversation content.

## Usage

When the user requests any of the following:
- "帮我总结一下这段对话"
- "生成会话小结"
- "对这些聊天记录做个摘要"
- "总结一下我们刚才聊了什么"

Use the `summarize_conversation` tool to call the summary API.

## Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| chat_list | string | Yes | JSON formatted conversation content |
| history_summary | string | No | Previous summary for incremental update |

### chat_list Format Example

```json
[
  {"role": "user", "content": "今天天气怎么样？"},
  {"role": "assistant", "content": "今天天气晴朗，气温25度。"}
]
```

## Response

The API returns JSON with:
- `code`: Status code, 0 means success
- `message`: Status message
- `data.summary`: Generated conversation summary

## Error Handling

- If the API returns a non-zero code, report the error message to the user
- If the request fails, check network connectivity
- Ensure chat_list is valid JSON format before calling