From faf
Access FAF documentation, guides, and resources. Answers questions about The Reading Order, IANA registration, Podium scoring, format specification, and best practices. Use when user asks "how does FAF work", "show me docs", "explain The Reading Order", or needs reference information.
How this skill is triggered — by the user, by Claude, or both
Slash command
/faf:faf-docsThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Provide quick access to FAF documentation, guides, specifications, and learning resources. Answers common questions and explains FAF concepts without requiring users to leave their workflow.
Provide quick access to FAF documentation, guides, specifications, and learning resources. Answers common questions and explains FAF concepts without requiring users to leave their workflow.
The Goal: Instant answers. Zero context switching. Championship-grade documentation.
This skill activates automatically when the user:
Trigger Words: docs, documentation, guide, explain, how does, what is, show me, reference, spec, help
Determine what the user needs:
Conceptual:
Reference:
Practical:
For conceptual questions:
For reference questions:
For practical questions:
After answering, provide:
What it is: Optimal sequence for AI to read project files for complete context understanding.
The Order:
1. project.faf → Project DNA (architecture, purpose, stack)
2. CLAUDE.md → Workflow instructions (how to work here)
3. README.md → User documentation (how humans use it)
4. package.json → Dependencies (what it needs)
5. Config files → Build settings (how to compile)
6. Code → Implementation (what it does)
Why this order:
Analogy: Like reading blueprints before examining bricks. You understand WHAT you're building before diving into HOW it's built.
What it is: FAF is officially registered with the Internet Assigned Numbers Authority as a recognized media type.
Details:
application/vnd.faf+yamlWhy it matters:
How to verify:
# In your project.faf
metadata:
iana_media_type: application/vnd.faf+yaml
What it is: AI-readiness measurement system (0-100%) showing how well AI can understand your project.
Tiers:
What gets scored:
How to check:
faf score
How to improve:
faf enhance
What it is: Bidirectional synchronization between project.faf and CLAUDE.md in <10ms (achieved: 8ms).
How it works:
When to use:
# After editing either file
faf bi-sync
Performance:
project.faf (Project DNA):
CLAUDE.md (Workflow Instructions):
Together: Complete AI context (architecture + workflow)
Minimum:
Recommended:
Best:
Update when:
Don't update for:
Think: "Would a new developer need to know this?" If yes, update project.faf.
Yes. FAF works with ALL AI tools:
Why: FAF is a text file in YAML format. Any AI can read YAML. No special integration required. Universal by design.
You can still benefit personally:
Team adoption is automatic:
name: project-name
purpose: What this project does
version: 1.0.0
metadata:
format_version: 3.0.0
iana_media_type: application/vnd.faf+yaml
architecture:
type: web-app | library | api | cli
language: TypeScript | Python | etc.
name: project-name
purpose: Detailed description of what this project does and why it exists
version: 1.0.0
metadata:
format_version: 3.0.0
iana_media_type: application/vnd.faf+yaml
created: 2025-11-02
last_updated: 2025-11-02
architecture:
type: web-app
language: TypeScript
framework: Next.js 14
pattern: Server-side rendering with App Router
description: |
Comprehensive architecture description explaining
the structure, patterns, and design decisions.
stack:
runtime: Node.js 18+
framework: Next.js 14 (App Router)
dependencies:
- react: "^18.2.0"
- typescript: "^5.3.3"
styling: Tailwind CSS
database: PostgreSQL 15 (Prisma ORM)
testing:
framework: Jest + React Testing Library
approach: |
- Unit tests for utilities
- Component tests for UI
- Integration tests for API routes
coverage: 78% overall, 90%+ critical paths
commands:
test: npm test
coverage: npm run test:coverage
build:
tool: Next.js (Turbopack)
commands:
dev: npm run dev
build: npm run build
start: npm start
deployment:
platform: Vercel
environment: Node.js 18
env_variables:
- DATABASE_URL
- JWT_SECRET
architecture_decisions:
- decision: "Chose App Router over Pages Router"
rationale: "Better performance, simpler data fetching"
- decision: "TypeScript strict mode"
rationale: "Catch errors at compile time"
performance:
- React Server Components for data fetching
- Image optimization with next/image
- Code splitting by route (automatic)
Essential:
faf init # Initialize project.faf
faf score # Check AI-readiness (0-100%)
faf enhance # Improve project.faf
faf validate # Check format compliance
faf bi-sync # Sync project.faf ↔ CLAUDE.md
faf migrate # Upgrade to latest format
Discovery:
faf auto # Auto-detect and generate everything
faf doctor # Health check for project
faf verify # Verify installation
Analysis:
faf formats # Show detected formats
faf stacks # Show detected tech stacks
faf vibe # AI-powered project analysis
Utility:
faf show # Display current project.faf
faf version # Show faf-cli version
faf help # Show help
Website:
GitHub:
npm:
Homebrew:
brew install wolfe-jam/faf/faf-cli
For specific tasks, use:
Answer:
"The Reading Order is the optimal sequence for AI to read your project files:
1. project.faf → Project DNA (WHAT it is)
2. CLAUDE.md → Workflow (HOW to work)
3. README.md → Docs (HOW humans use it)
4. package.json → Dependencies
5. Config files → Build settings
6. Code → Implementation
Why this order matters:
Analogy: Reading blueprints before examining bricks. You understand the building plan before diving into construction details.
Result: AI has complete context in <1 second vs 10-30 minutes of questions."
Answer:
"Here's the IANA-registered format specification:
Minimal valid project.faf:
name: project-name
purpose: What it does
version: 1.0.0
metadata:
format_version: 3.0.0
iana_media_type: application/vnd.faf+yaml
architecture:
type: web-app
language: TypeScript
Create yours:
faf init # Auto-generates valid format
Full spec: https://faf.one/docs/format"
Answer:
"Use the faf-enhance skill for guided improvement:
faf enhance
This will:
Typical progression:
Want me to activate the faf-enhance skill now?"
Documentation as Code:
NO BS ZONE:
Format-Driven:
When this skill succeeds, users should:
Generated by FAF Skill: faf-docs v1.0.0 Documentation Edition: Instant Answers "The Reading Order. IANA registration. Championship docs."
npx claudepluginhub wolfe-jam/faf-skillsExpert in the .faf format for persistent project context across AI tools. Scores AI-readiness, syncs context to CLAUDE.md, and manages the faf-cli toolkit. Use when working with .faf files or the FCL doctrine.
Configures IANA-registered .faf files for persistent AI project context, MCP server setup, championship scoring, and bi-directional sync across Claude, Cursor, Gemini, Windsurf.
Guides project documentation structure, content requirements, and best practices. Includes templates for README, ARCHITECTURE, API docs, and change history.