Lobster Compact

针对接近上下文上限的长会话，按时间顺序分析每次交互，提取用户意图、技术概念、文件与代码、错误修复、用户反馈、待办与下一步，输出固定八章节的 Markdown 会话摘要，并将关键长期信息写入 MEMORY.md、当日日志写入 memory/YYYY-MM-DD.md，借助 memory_search 去重。

wangxiaofei860208-source

热度 46.83

最近更新 2026-05-11 17:40:07

分类：效率依赖：读写文件 / 内部调用大模型应用场景：摘要总结 / 笔记整理 / 工作流

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add lobster-compact

Metadata

name: lobster-compact
description: Summarize long conversations to preserve context. Automatically triggered when context window approaches limits, or manually with /compact.
allowed-tools: ["read","write","edit","memory_search","memory_get"]
when_to_use: Use when the conversation becomes very long and context may be lost. Examples: 'summarize so far', 'compact', '我们聊了太多了', '总结一下'.

Compact — 长会话压缩

Inputs

$args: Optional focus area for the summary

Goal

Create a detailed summary of the conversation so far that preserves all critical context for continuing work.

Steps

1. Analyze the Conversation

Chronologically analyze each exchange:

User's explicit requests and intents
Approach taken to address requests
Key decisions, technical concepts, code patterns
Specific details: file names, code snippets, function signatures
Errors encountered and how they were fixed
User feedback and corrections

2. Generate Summary

Write a summary with these sections:

# 会话摘要

## 1. 主要请求和意图
[用户的核心需求]

## 2. 关键技术概念
[涉及的技术、框架、工具]

## 3. 文件和代码
[涉及的文件和关键代码片段]

## 4. 错误和修复
[遇到的问题及解决方案]

## 5. 用户反馈
[用户给出的指导、纠正、偏好]

## 6. 待办任务
[尚未完成的任务]

## 7. 当前工作
[最近在做什么，精确到文件和代码]

## 8. 下一步
[紧接着应该做什么]

3. Update Memory

Save critical long-term info to MEMORY.md
Save today's log to memory/YYYY-MM-DD.md
Use memory_search to avoid duplicates

Success criteria: Summary captures all critical context; memory files updated.