💡 摘要
一组用于增强Obsidian和Markdown工作流程的个人知识管理的AI技能。
🎯 适合人群
🤖 AI 吐槽: “看起来很能打,但别让配置把人劝退。”
风险:Medium。建议检查:是否执行 shell/命令行指令;是否发起外网请求(SSRF/数据外发);API Key/Token 的获取、存储与泄露风险;文件读写范围与路径穿越风险;依赖锁定与供应链风险。以最小权限运行,并在生产环境启用前审计代码与依赖。
Claude Obsidian Skills
Reusable AI agent skills for Obsidian and Markdown workflows with Claude Code, Gemini CLI, and other AI assistants.
Overview
Claude Obsidian Skills is a collection of AI agent skills designed to automate personal knowledge management (PKM) workflows. These skills work with Claude Code, Gemini CLI, and other AI assistants that support skill-based extensions.
Available Skills
PKM Management Skills
| Skill | Description | |-------|-------------| | ai4pkm-helper | AI4PKM helper for onboarding guidance, quick help, and seamless handoff to DDA for daily use |
Obsidian Skills
| Skill | Description | |-------|-------------| | obsidian-canvas | Create and manage Obsidian Canvas files with automatic layout generation for knowledge maps and visual summaries | | obsidian-links | Format, validate, and fix wiki links with proper filename, section, and folder conventions | | obsidian-yaml-frontmatter | Standardize YAML frontmatter properties with consistent formatting and naming | | obsidian-markdown-structure | Validate and enforce markdown document structure, heading hierarchy, and organization | | obsidian-mermaid | Create Obsidian-compatible Mermaid diagrams avoiding common errors (markdown in labels, wide layouts) |
Markdown Skills
| Skill | Description | |-------|-------------| | markdown-slides | Create Deckset/Marp compatible presentation slides from markdown content | | interactive-writing-assistant | AI-powered writing companion with outline-prose co-evolution and PKM integration |
Video Skills
| Skill | Description | |-------|-------------| | markdown-video | Convert markdown slides to MP4 video with TTS narration (with or without Deckset) | | video-cleaning | Remove silent pauses from videos using AI-powered transcription and FFmpeg |
Image Skills
| Skill | Description | |-------|-------------| | gemini-image-skill | Generate AI images using Google Gemini API with multiple models and aspect ratios |
Installation
For Claude Code
- Clone this repository or copy the skill folders:
git clone https://github.com/jykim/claude-obsidian-skills.git
- Copy skills to your Claude Code skills directory:
cp -r claude-obsidian-skills/*-* ~/.claude/skills/
- Claude Code will automatically detect and use the skills.
For Other AI Assistants
Copy the SKILL.md file from the desired skill folder to your assistant's skill directory. Each skill is self-contained and documented within its SKILL.md file.
Advanced: Git Submodule Integration
For vault maintainers who want to keep skills in sync with this repository while avoiding duplication:
- Add as a submodule to your vault:
git submodule add https://github.com/jykim/claude-obsidian-skills.git path/to/skills/public-skills
- Create symlinks for backward compatibility:
cd path/to/skills ln -s public-skills/obsidian-canvas obsidian-canvas ln -s public-skills/obsidian-links obsidian-links ln -s public-skills/obsidian-yaml-frontmatter obsidian-yaml-frontmatter ln -s public-skills/obsidian-markdown-structure obsidian-markdown-structure ln -s public-skills/obsidian-mermaid obsidian-mermaid ln -s public-skills/markdown-slides markdown-slides ln -s public-skills/markdown-video markdown-video ln -s public-skills/interactive-writing-assistant interactive-writing-assistant ln -s public-skills/video-cleaning video-cleaning
- Update skills when new versions are released:
git submodule update --remote
Benefits:
- Single source of truth - edit in one place
- Version control - pin to specific commits if needed
- Easy updates via git submodule commands
Note: When cloning a vault with submodules, use --recursive:
git clone --recursive your-vault-repo
Requirements
Some skills require external dependencies. Install them before using:
| Skill | Requirements | |-------|--------------| | markdown-video | FFmpeg, OpenAI API key, Python 3.7+, Pillow | | video-cleaning | FFmpeg, OpenAI API key, Python 3.7+ | | gemini-image-skill | Gemini API key, Python 3.7+, google-genai, Pillow | | markdown-slides | None (markdown only) | | obsidian-* | None (file operations only) | | interactive-writing-assistant | None |
Common Installation Commands
# FFmpeg (macOS) brew install ffmpeg # Python dependencies pip install Pillow openai google-genai # Set API keys export OPENAI_API_KEY="sk-..." export GEMINI_API_KEY="your-gemini-api-key"
Quick Start
Validate Wiki Links
Use the obsidian-links skill to check and fix broken wiki links in my vault
Standardize Frontmatter
Use obsidian-yaml-frontmatter to ensure consistent properties across my notes
Create Mermaid Diagrams
Use obsidian-mermaid to create a flowchart for this process
Create Visual Canvas
Use obsidian-canvas to create a knowledge map for this topic
Create Presentation
Use markdown-slides to convert this document into a Deckset presentation
Generate Video
Use markdown-video to create a narrated video from my slides
Clean Video
Use video-cleaning to remove silent pauses from this video
Skill Structure
Each skill follows a consistent structure:
skill-name/
├── SKILL.md # Main skill definition and documentation
├── reference/ # Reference materials (optional)
└── examples/ # Usage examples (optional)
The SKILL.md file contains:
- Skill metadata (name, description, allowed tools)
- When to use the skill
- Core rules and guidelines
- Step-by-step workflows
- Quality checklists
Contributing
We welcome contributions! Here's how you can help:
- Report bugs - Open an issue describing the problem
- Suggest improvements - Share ideas for enhancing existing skills
- Add new skills - Create skills for other PKM workflows
- Improve documentation - Help clarify usage and examples
Creating a New Skill
- Create a folder in the repo root with your skill name
- Add a
SKILL.mdwith:- Frontmatter (name, description, allowed-tools, license)
- Clear "When to Use" section
- Step-by-step process documentation
- Quality checklist
- Test the skill with Claude Code or your preferred assistant
- Submit a pull request
Related Projects
- AI4PKM - Full PKM automation framework
- AI4PKM Docs - Documentation and guides
- Discord Community - Join the discussion
AI4PKM Framework
These skills are part of the AI4PKM framework for AI-powered personal knowledge management. For the complete automation system including workflows, templates, and integrations, check out the main project.
License
MIT License - see LICENSE for details.
Made with AI assistance for AI assistance.
优点
- 适用于各种任务的广泛技能
- 与现有工作流程良好集成
- 支持多种AI助手
缺点
- 某些技能需要外部依赖
- 新用户的学习曲线
- 文档可以更详细
相关技能
免责声明:本内容来源于 GitHub 开源项目,仅供展示和评分分析使用。
版权归原作者所有 jykim.
