Back to Skills

documentation-standards

pr-pm
Updated Today
26 views
62
9
62
View on GitHub
Designwordaidesign

About

This skill provides documentation standards for organizing and maintaining PRPM repository docs. It defines clear structures for user-facing documentation (`docs/`) and internal developer documentation (`development/docs/`). Use it to ensure consistency across all documentation types when contributing to or using the PRPM project.

Quick Install

Claude Code

Recommended
Plugin CommandRecommended
/plugin add https://github.com/pr-pm/prpm
Git CloneAlternative
git clone https://github.com/pr-pm/prpm.git ~/.claude/skills/documentation-standards

Copy and paste this command in Claude Code to install this skill

Documentation

PRPM Documentation Standards

Documentation Organization

Internal Documentation (development/docs/)

Purpose: Documentation for developers working on PRPM itself

Location: development/docs/

Files:

  • GITHUB_WORKFLOWS.md - GitHub Actions workflows reference
  • PUBLISHING.md - NPM package publishing process and order
  • DEVELOPMENT.md - Development setup, environment, and workflows
  • DOCKER.md - Docker setup, services, and troubleshooting

Audience: PRPM contributors, maintainers, CI/CD systems


User-Facing Documentation (docs/)

Purpose: Documentation for PRPM users and package authors

Location: docs/ (at project root)

Files:

  • User guides
  • API documentation
  • Package authoring guides
  • CLI command reference
  • Examples and tutorials

Audience: PRPM end users, package authors, integrators


Project-Level Documentation (root)

Purpose: Standard project files that belong at repository root

Location: Project root /

Files:

  • README.md - Project overview, quick start, installation
  • CONTRIBUTING.md - Contribution guidelines
  • CHANGELOG.md - Version history and changes
  • LICENSE - License information
  • ROADMAP.md - Project roadmap and future plans

Audience: Everyone (first impression)


Claude Skills (.claude/skills/)

Purpose: Knowledge base and reference materials for AI assistants

Location: .claude/skills/

Files:

  • postgres-migrations-skill.md - PostgreSQL migrations guidance
  • pulumi-troubleshooting-skill.md - Pulumi troubleshooting
  • NEW_SKILLS.md - How to create new skills
  • documentation-standards.md - This file

Subdirectories:

  • prpm-development/ - PRPM-specific development knowledge
  • self-improving/ - Self-improvement patterns
  • thoroughness/ - Thoroughness and quality guidelines

Audience: AI assistants (Claude, etc.)


Rules for Documentation Placement

When to use development/docs/

✅ GitHub Actions workflows and CI/CD ✅ Internal build/release processes ✅ Development environment setup ✅ Architecture decision records ✅ Internal troubleshooting guides ✅ Database schema documentation ✅ Infrastructure documentation

❌ User-facing tutorials ❌ CLI usage guides ❌ API reference for end users

When to use docs/

✅ User guides and tutorials ✅ CLI command reference ✅ Package authoring guides ✅ API documentation for users ✅ Integration examples ✅ FAQ for users

❌ Internal development workflows ❌ CI/CD documentation ❌ Build/release processes

When to use .claude/skills/

✅ Specialized knowledge for AI assistants ✅ Domain-specific best practices ✅ Troubleshooting patterns ✅ Code review guidelines ✅ Project-specific conventions

❌ General documentation ❌ User guides ❌ API references


Documentation Standards

Markdown Files

  • Use clear, descriptive filenames (kebab-case)
  • Include table of contents for docs > 200 lines
  • Use proper heading hierarchy (# → ## → ###)
  • Include code examples with syntax highlighting
  • Add frontmatter if using a static site generator

Example Structure

# Title

Brief description (1-2 sentences)

## Table of Contents
- [Section 1](#section-1)
- [Section 2](#section-2)

## Section 1
Content...

### Subsection 1.1
Details...

## Examples
\`\`\`bash
# Example command
prpm install @username/package-name
\`\`\`

## See Also
- [Related Doc](./related.md)

Cross-References

  • Use relative paths for links
  • Keep links within same category when possible
  • Update links when moving files

Internal → Internal:

See [Publishing Guide](./PUBLISHING.md)

Internal → User:

See [User Guide](../../docs/user-guide.md)

Migration Checklist

When reorganizing documentation:

  1. ✅ Move file to correct location
  2. ✅ Update all references to moved file
  3. ✅ Update README.md links if needed
  4. ✅ Update .gitignore if needed
  5. ✅ Test that all links work
  6. ✅ Commit with clear message explaining move

Package-Specific Documentation

Each package should have its own README:

packages/
├── cli/
│   └── README.md          # CLI package overview
├── registry/
│   └── README.md          # Registry server docs
├── registry-client/
│   └── README.md          # Client library docs
├── types/
│   └── README.md          # Type definitions docs
└── webapp/
    └── README.md          # WebApp docs

Maintenance

Regular Reviews

  • Quarterly review of docs/ for accuracy
  • Remove outdated documentation
  • Update examples to use latest version
  • Check for broken links

When Adding Features

  • Update relevant user docs in docs/
  • Update internal docs in development/docs/ if needed
  • Add examples
  • Update CHANGELOG.md

When Deprecating Features

  • Add deprecation notice to docs
  • Provide migration guide
  • Keep docs until feature is removed
  • Update CHANGELOG.md

Quick Reference

Documentation TypeLocationAudienceExamples
Internal Devdevelopment/docs/ContributorsCI/CD, publishing
User-Facingdocs/UsersGuides, tutorials
Project Root/EveryoneREADME, LICENSE
AI Skills.claude/skills/AI assistantsTroubleshooting
Package Docspackages/*/README.mdPackage usersAPI reference

Tools

Documentation Generators

  • TypeDoc - For TypeScript API docs (future)
  • VitePress or Docusaurus - For docs/ site (future)

Linting

# Check markdown
markdownlint docs/

# Check links
markdown-link-check docs/**/*.md

Building Docs Site (Future)

cd docs/
npm run build

Examples

Good Documentation Structure

prpm/
├── README.md                    # Project overview
├── CONTRIBUTING.md              # How to contribute
├── CHANGELOG.md                 # Version history
├── ROADMAP.md                   # Future plans
├── development/
│   └── docs/
│       ├── GITHUB_WORKFLOWS.md  # CI/CD reference
│       ├── PUBLISHING.md        # Release process
│       ├── DEVELOPMENT.md       # Dev setup
│       └── DOCKER.md            # Services setup
├── docs/
│   ├── getting-started.md       # User onboarding
│   ├── cli-reference.md         # Command reference
│   ├── package-authoring.md     # Creating packages
│   └── api/
│       └── registry-client.md   # API docs
└── .claude/
    └── skills/
        ├── documentation-standards.md
        ├── postgres-migrations-skill.md
        └── pulumi-troubleshooting-skill.md

Bad Documentation Structure ❌

prpm/
├── README.md
├── WORKFLOWS.md                 # Should be in development/docs/
├── USER_GUIDE.md                # Should be in docs/
├── dev-setup.md                 # Should be in development/docs/
└── troubleshooting.md           # Unclear audience/location

Version

Last Updated: 2025-10-21 Applies To: PRPM v2+ Review Date: 2026-01-21

GitHub Repository

pr-pm/prpm
Path: .claude/skills/documentation-standards
claudeclaude-codecursorcursor-ai-editcursorrulespackage-manager

Related Skills

sglang

Meta

SGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.

View skill

evaluating-llms-harness

Testing

This Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.

View skill

llamaguard

Other

LlamaGuard is Meta's 7-8B parameter model for moderating LLM inputs and outputs across six safety categories like violence and hate speech. It offers 94-95% accuracy and can be deployed using vLLM, Hugging Face, or Amazon SageMaker. Use this skill to easily integrate content filtering and safety guardrails into your AI applications.

View skill

langchain

Meta

LangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.

View skill