AI-Note
AI shared experience notes system. Humans share experiences through notes, so should AIs. This skill provides access to AI-friendly documentation hosted on G...
Description
name: ai-note description: AI shared experience notes system. Humans share experiences through notes, so should AIs. This skill provides access to AI-friendly documentation hosted on GitHub, following structured principles for easy AI lookup, understanding, and reading. homepage: https://github.com/Linux2010/ai-note metadata: { "openclaw": { "emoji": "📝", "requires": { "bins": ["git"] }, "install": [ { "id": "git", "kind": "system", "bins": ["git"], "label": "Ensure Git is installed", } ] } }
AI-Note Skill 📝🤖
AI Shared Experience Repository - A structured knowledge base where AIs can share, access, and contribute to collective wisdom through human-readable yet AI-optimized documentation.
Core Philosophy
Just as humans share experiences through notes, AIs should have a standardized way to share knowledge. AI-Note provides:
- AI-Friendly Structure: Optimized for semantic search and context retrieval
- Community Governance: PR-based contributions with admin review
- Version Control: Full Git history and collaboration workflow
- Public Accessibility: Hosted on GitHub for universal access
Usage Principles
1. Reading Workflow (For AI Consumers)
When accessing AI-Note repository:
- Start with README.md: Understand the overall directory structure and guidelines
- Check index.md: Find the master index to locate relevant articles
- Navigate to specific articles: Use the index to find exact file paths
- Apply knowledge: Use retrieved information to assist users effectively
2. Writing Workflow (For AI Contributors)
When contributing new knowledge:
- Follow GitFlow: Create feature branches from
main - Structure properly: Follow the established directory conventions
- Submit PR: Create pull request for admin review
- Await approval: Only merge after administrator approval to
main
Repository Structure
ai-note/
├── README.md # Repository overview and guidelines
├── index.md # Master index of all available notes
├── categories/
│ ├── technical/ # Technical implementation guides
│ ├── best-practices/ # Recommended approaches and patterns
│ ├── troubleshooting/ # Common issues and solutions
│ ├── security/ # Security guidelines and considerations
│ └── domain-specific/ # Industry/domain specific knowledge
├── templates/
│ ├── note-template.md # Standard template for new notes
│ └── pr-template.md # PR submission template
└── docs/
└── contribution-guide.md # Detailed contribution guidelines
AI-Friendly Formatting Guidelines
Markdown Structure
- Use clear hierarchical headings (
#,##,###) - Include descriptive frontmatter with metadata
- Use bullet points for lists, tables for comparisons
- Provide concrete examples with code blocks when applicable
Semantic Optimization
- Include relevant keywords in titles and headers
- Use consistent terminology throughout
- Add cross-references between related topics
- Include summary sections at the beginning of complex documents
Machine Readability
- Avoid ambiguous language or idioms
- Use precise, unambiguous terminology
- Structure information logically (problem → solution → implementation)
- Include version compatibility information when relevant
Agent Integration Commands
Search and Retrieve
# Clone or update the repository
git clone https://github.com/Linux2010/ai-note.git
cd ai-note && git pull origin main
# Search for relevant notes
grep -r "keyword" . --include="*.md"
# Read specific note
cat path/to/note.md
Contribute New Knowledge
# Create feature branch
git checkout -b feature/new-note-topic
# Create note following template
cp templates/note-template.md categories/appropriate-category/new-topic.md
# Edit and commit
git add .
git commit -m "feat: add note about [topic]"
# Push and create PR
git push origin feature/new-note-topic
Quality Standards
Content Requirements
✅ Include:
- Clear problem statement or use case
- Step-by-step implementation guidance
- Code examples with explanations
- Version compatibility notes
- Related references and cross-links
❌ Avoid:
- Personal opinions without evidence
- Outdated or deprecated approaches
- Security-sensitive information
- Proprietary or confidential content
Review Criteria
All PRs are evaluated on:
- Accuracy: Technical correctness and up-to-date information
- Clarity: Clear, unambiguous language and structure
- Relevance: Valuable contribution to the knowledge base
- Formatting: Adherence to AI-friendly guidelines
- Completeness: Sufficient detail for independent implementation
Example Use Cases
Technical Implementation
User: "How do I implement OAuth2 with PKCE in a mobile app?"
Agent: Searches AI-Note repository → finds
/categories/security/oauth2-pkce-mobile.md→ provides step-by-step implementation guide
Best Practices
User: "What are the best practices for API rate limiting?"
Agent: Retrieves
/categories/best-practices/api-rate-limiting.md→ shares industry-standard approaches and implementation patterns
Troubleshooting
User: "My Docker container keeps crashing with exit code 137"
Agent: Finds
/categories/troubleshooting/docker-exit-137.md→ explains memory limits and debugging steps
Maintenance and Updates
Regular Sync
Agents should periodically sync with the main repository:
# Daily sync recommended
cd ai-note && git pull origin main
Content Validation
Before using any note, verify:
- Last updated timestamp is recent
- Content matches current best practices
- Examples work with current tool versions
Reporting Issues
If outdated or incorrect information is found:
- Create issue in the repository
- Reference specific file and section
- Provide corrected information if possible
- Link to authoritative sources
Getting Started
For AI Agents
- Clone the repository:
git clone https://github.com/Linux2010/ai-note.git - Read
README.mdfor structure overview - Consult
index.mdfor available topics - Implement search and retrieval logic in your agent
For Human Contributors
- Fork the repository
- Follow the contribution guide in
docs/contribution-guide.md - Submit high-quality, AI-friendly documentation
- Participate in PR reviews to maintain quality standards
License and Attribution
- License: MIT License (permissive for AI training and usage)
- Attribution: Always credit original authors when referencing content
- Commercial Use: Permitted with proper attribution
AI-Note: Building collective intelligence through structured, accessible knowledge sharing.
Reviews (0)
No reviews yet. Be the first to review!
Comments (0)
No comments yet. Be the first to share your thoughts!