메뉴
从当前对话中提取知识到用户指定的知识库。仅在用户显式要求保存知识、沉淀经验、更新 wiki、capture knowledge、preserve knowledge 时使用。
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a talk/pitch. Helps non-designers discover their aesthetic through visual exploration rather than abstract choices.
Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a talk/pitch. Helps non-designers discover their aesthetic through visual exploration rather than abstract choices.
API Key penetration testing skill for buyer/seller platform API key security assessment. Runs secret scanning (trufflehog + gitleaks), endpoint fuzzing (ffuf + arjun), traffic analysis, and records all findings.
Autonomous research review loop using any OpenAI-compatible LLM API. Configure via llm-chat MCP server or environment variables. Trigger with "auto review loop llm" or "llm review".
Autonomous multi-round research review loop using MiniMax API. Use when you want to use MiniMax instead of Codex MCP for external review. Trigger with "auto review loop minimax" or "minimax review".
Autonomous multi-round research review loop. Repeatedly reviews via Codex MCP, implements fixes, and re-reviews until positive assessment or max rounds reached. Use when user says "auto review loop", "review until it passes", or wants autonomous iterative improvement.
| name | knowledge-preserve |
| description | 从当前对话中提取知识到用户指定的知识库。仅在用户显式要求保存知识、沉淀经验、更新 wiki、capture knowledge、preserve knowledge 时使用。 |
仅在用户显式要求保存知识时使用。不要在普通对话结束后自动写入知识库。
优先使用用户在本轮提供的知识库路径。若用户没有提供路径，询问目标位置，不要假设固定目录。
可使用的通用约定：
| 用途 | 示例路径 |
|---|---|
| 团队/项目知识库 | ~/KnowledgeBase/Work |
| 个人知识库 | ~/KnowledgeBase/Personal |
/knowledge-preserve # 默认：完整原文存 raw + 自动消化到 wiki
/knowledge-preserve direct # 有隐私：raw 只存摘要，但消化完整原文到 wiki
第一步永远是确认输入人。
如目标知识库内存在开发规范、踩坑记录、ADR 或模板文件，先读取这些项目内文档。
两种模式都会：
区别只在 raw 里存什么：
确认输入人 → 完整原文存 raw → 消化完整原文 → 审核+矛盾+重复 → 写 wiki
<KNOWLEDGE_BASE>/原始资料/对话记录/{输入人}/{date}-{topic}.mdcompiled: true---
type: conversation-extract
输入人: will
会话开始: 2026-04-12 14:30
会话结束: 2026-04-12 16:45
来源目录: __HOME__/projects/some-project
topics: [CouchDB配置, CORS问题, LiveSync]
compiled: true
tags: [conversation, will]
---
# 2026-04-12 CouchDB配置和LiveSync部署
## 14:30 — 开始配置 CouchDB
（完整对话原文...）
## 15:20 — 遇到 CORS 问题
（完整对话原文...）
## 16:30 — LiveSync 连接成功
（完整对话原文...）
如果同一个 CC 会话窗口多次执行 /knowledge-preserve：
会话结束 时间戳，会话开始 不变确认输入人 → 消化完整原文 → 过滤隐私 → 审核+矛盾+重复 → 写 wiki → raw 存摘要
原始资料/对话记录/{输入人}/{date}-{topic}-摘要.md