packages/thinking-framework-skills/skills/create-topic/SKILL.md
# Create Topic Node ## Goal Create a new topic (or sub-topic) directory in the knowledge tree with basic node initialization. Includes automatic best placement discovery, user interaction for confirmation, duplicate checking, and directory tree output. ## Input - Parent directory path (passed from SKILL.md after config check) - New topic name - One-sentence topic definition - Optional: initial sub-topic list ## Execution Steps ### 1. Find Best Placement Directory Invoke the **[Retrieve To
npx skillsauth add edisonlzy/ai-codemons packages/thinking-framework-skills/skills/create-topicInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Create a new topic (or sub-topic) directory in the knowledge tree with basic node initialization. Includes automatic best placement discovery, user interaction for confirmation, duplicate checking, and directory tree output.
Invoke the Retrieve Topic Node subskill to locate the most suitable parent directory in the knowledge tree.
⚠️ Warning: Directory [parent_path]/[topic_name] already exists
Existing content:
- README.md
- FAQ.md
- 1.知识文档A.md
- 2.知识文档B.md
- subtopicA/
Overwrite? [y/N]
Create the following directory structure:
/[topic-name]/
├── README.md # 必须:Definition + Boundary + Sub-topic Index
├── FAQ.md # 必须:常见问题 + 实践经验沉淀
├── 1.知识文档A.md # 可选:知识文档,带序号前缀
├── 2.知识文档B.md # 可选:知识文档,带序号前缀
└── [subtopic1]/ # 可选:子 Topic
├── README.md
├── FAQ.md
└── 1.子topic知识文档.md # 可选:子 Topic 下的知识文档
⚠️ 约束:如果需要新创建文件,只需要参考模版生成文件的内容结构即可,不要乱加任何未提及的内容。
Use resources/README-template.md as the template. Only fill in content provided by the user - leave placeholders for content not mentioned (e.g., Sub-topic Index section should remain with placeholder content).
⚠️ 约束:如果需要新创建文件,只需要参考模版生成文件的内容结构即可,不要乱加任何未提及的内容。
Use resources/FAQ-template.md as the template, copy the template content and add the scope description. Only include content provided by the user.
⚠️ 约束:FAQ 中的 callout 块(如
> [!faq]-)标题行后必须有空行>,否则无法正确渲染为可折叠块。添加问题时确保每个部分之间有空行分隔。
This step is REQUIRED. Follow the complete workflow defined in Optimize Topic Files.
After creating the directory, update the parent directory's README.md file to add sub-topic index:
README.md## Sub-topic Index:## Sub-topic Index
- [[topic_name/README.md]]
If ## Sub-topic Index section does not exist, create it.
If the user provides an initial sub-topic list, recursively create sub-topic directories with their README.md and FAQ.md.
Use the tree command or manual traversal to output the created directory structure:
✅ Directory creation complete!
[parent_directory]/
└── [topic_name]/
├── README.md
├── FAQ.md
├── 1.知识文档A.md ← (optional, if provided)
├── 2.知识文档B.md ← (optional, if provided)
└── [subtopic1]/
├── README.md
├── FAQ.md
└── 1.子topic文档.md ← (optional, if provided)
README.md and FAQ.md, optionally with numbered knowledge docs)README.md follows three-section format (Definition + Boundary + Sub-topic Index)FAQ.md uses template structureREADME.md updated with new topic in sub-topic indextree -L 3 <path> commanddevelopment
# Retrieve Topic Node ## Goal Locate the most relevant topic directory in the knowledge tree based on the user's request. Following the progressive disclosure principle, traversal starts from the vault root and descends layer by layer, with user confirmation at each proposed match. ## Runtime Environment 本 Skill 由主 Agent 调用,dispatch 一个 **general-purpose subagent** 来执行实际的检索操作。 - 主 Agent 负责任务分发和结果聚合 - Subagent 执行目录遍历和关键词匹配 - Subagent 仅返回检索结果,不做决策 ## Input - Vault root path (resolved from con
development
Apply Obsidian-flavoured formatting to a topic node's README.md, FAQ.md, and content knowledge-point files using standard templates.
development
Locate the most relevant topic directory in the Obsidian knowledge tree based on a search query.
development
Divide a large topic into smaller, well-defined sub-topic nodes through web research and user collaboration.