菜单
Guides the creation of effective Claude Code skills with proper structure, metadata, and progressive disclosure. Use when creating new skills, refactoring existing skills, or reviewing skill effectiveness for Claude Code projects.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Provides deep architectural insights into Topcat's implementation including module structure, DAG algorithms, layer system, analysis capabilities, and schema operations. Use when exploring Topcat internals, understanding algorithms, extending functionality, or debugging complex issues.
Guides graph export and visualization in multiple formats. Use when visualizing dependencies, generating documentation, integrating with analysis tools, or creating diagrams for reports.
Guides schema-based organization and analysis for multi-schema database projects. Use when working with multiple schemas, analyzing cross-schema dependencies, or filtering operations by schema.
Guides dependency analysis and safe cleanup operations in Topcat. Use when analyzing dead branches, finding orphans, detecting cycles, cleaning unused files, or integrating with CI/CD pipelines.
Guides testing and debugging of Topcat including running tests, writing new test cases, and debugging issues. Use when working with Topcat's test suite, testing analysis commands, debugging dependency problems, analyzing performance, or setting up continuous integration.
Systematically runs clippy and fixes linting issues following project patterns. Use when running lints, addressing clippy warnings, or improving code quality before commits.
| name | writing-skills |
| description | Guides the creation of effective Claude Code skills with proper structure, metadata, and progressive disclosure. Use when creating new skills, refactoring existing skills, or reviewing skill effectiveness for Claude Code projects. |
Create a new skill following this workflow:
Skill Creation Checklist:
- [ ] Step 1: Identify the reusable pattern
- [ ] Step 2: Create skill directory with SKILL.md
- [ ] Step 3: Write YAML frontmatter (name, description)
- [ ] Step 4: Draft concise main content (<200 lines)
- [ ] Step 5: Extract details to reference files
- [ ] Step 6: Test with real usage scenarios
- [ ] Step 7: Iterate based on Claude's behavior
skill-name/
├── SKILL.md # Main file (loaded when triggered)
├── reference/ # Detailed guides (loaded as needed)
│ ├── guide1.md
│ └── guide2.md
└── scripts/ # Utility scripts (executed, not loaded)
└── helper.py
---
name: skill-name # lowercase, hyphens only, max 64 chars
description: What this skill does and when to use it. # max 1024 chars
---
## Task Workflow
Copy this checklist:
\`\`\`
Progress:
- [ ] Step 1: Clear action
- [ ] Step 2: Next action
- [ ] Step 3: Final action
\`\`\`
## Quick Guide
[Essential information here]
**Details**: See [reference/details.md](reference/details.md)
**Examples**: See [reference/examples.md](reference/examples.md)
For templates, see templates/. For best practices, see reference/best-practices.md. For anti-patterns, see reference/anti-patterns.md.
Work with Claude to refine skills:
Iteration Checklist:
- [ ] Complete task without skill (note what context you provide)
- [ ] Create skill capturing that context
- [ ] Test skill with fresh Claude instance
- [ ] Observe behavior and note issues
- [ ] Refine based on observations
- [ ] Repeat until effective