From fh
Display the fhhs-skills command reference and architecture guide. Use when the user says 'help', 'what commands', 'show commands', 'how do composites work', 'skills guide', or needs an overview of available commands and how they fit together.
How this skill is triggered — by the user, by Claude, or both
Slash command
/fh:helpThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Output ONLY the reference content below. Do NOT add project-specific analysis, git status, next-step suggestions, or any commentary.
Output ONLY the reference content below. Do NOT add project-specific analysis, git status, next-step suggestions, or any commentary.
# fhhs-skills Command ReferenceFirst time? Follow these steps:
/fh:setup — Install tools, configure LSP, set up hooks (~2 min, once)/fh:new-project — Define your project vision, tech stack, and roadmap/fh:plan — Plan your first feature or phase/fh:build — Execute the plan with tests, quality checks, and verificationReturning? Run /fh:progress to see where you left off and what's next.
Need help? You're reading it. See command tables below, or ask any question about a specific command.
| Command | What it does |
|---|---|
/fh:new-project | Set up a project with vision, tech stack, design language, and roadmap |
/fh:plan | Brainstorm, research, and produce an execution-ready plan. Handles phase creation and placement automatically |
/fh:build | Execute a plan with parallel subagents, TDD, design gates, and verification. Handles milestone completion automatically |
/fh:verify | Goal-backward verification with truth tables and fresh evidence |
/fh:verify-ui | Visual verification with browser screenshots |
| Command | What it does |
|---|---|
/fh:fix | Auto-triage and fix bugs with systematic debugging |
/fh:refactor | Restructure code safely, tests green at every step |
/fh:research | Investigate a topic before planning |
| Command | What it does |
|---|---|
/fh:resume | Restore context and route to the right next action |
/fh:progress | See where you are and what's next |
| Command | What it does |
|---|---|
/fh:critique | Evaluate visual hierarchy, information architecture, and design quality |
/fh:ui-animate | Purposeful motion and micro-interactions |
/fh:ui-branding | One-time setup for your project's design language |
Design agents (dispatched automatically by /fh:review, /fh:refactor, /fh:fix, and /fh:build):
| Agent | What it does |
|---|---|
fh:design-polish | Fix alignment, spacing, consistency, and detail issues |
fh:design-normalize | Match your design system and ensure consistency |
fh:design-harden | Error handling, i18n, text overflow, edge cases |
fh:design-audit | Full accessibility, performance, theming, and responsive audit |
fh:design-adapt | Make designs work across screen sizes and platforms |
fh:design-bolder | Amplify safe designs to be more visually interesting |
fh:design-quieter | Tone down overly aggressive designs |
fh:design-distill | Strip away unnecessary complexity |
fh:design-clarify | Improve confusing labels, errors, and microcopy |
fh:design-colorize | Add strategic color to monochromatic interfaces |
fh:design-delight | Add personality and moments of joy |
fh:design-extract | Pull reusable components into your design system |
fh:design-onboard | Design first-time user experiences and empty states |
fh:design-optimize | Improve loading speed, rendering, and bundle size |
fh:design-secure | Security scanning of changed files |
fh:design-simplify | Review code for reuse, quality, and efficiency |
| Command | What it does |
|---|---|
/fh:todos | Track things to do — add or review pending todos |
Phase management, milestone lifecycle, and test generation are handled automatically by /fh:plan, /fh:build, and /fh:verify.
| Command | What it does |
|---|---|
/fh:setup | One-time setup after installing |
/fh:settings | Configure workflow preferences |
/fh:health | Check if your project files are in good shape |
/fh:map-codebase | Analyze your codebase structure |
/fh:learnings | Capture and apply session learnings, optionally update CLAUDE.md |
/fh:update | Check for updates and install the latest version |
First time: See "Getting Started" above
Feature: /fh:plan → /fh:build → /fh:verify → /fh:verify-ui
Bug fix: /fh:fix
Refactoring: /fh:refactor (dispatches fh:design-simplify agent automatically)
Resuming: /fh:resume → (routes to next action)
CLAUDE.md: /fh:learnings (after sessions to capture and apply learnings)
.planning/
├── PROJECT.md # Project vision
├── ROADMAP.md # Current phase breakdown
├── STATE.md # Project memory & context
├── DESIGN.md # Design language (from /fh:ui-branding)
├── config.json # Workflow mode & gates
├── todos/ # Captured ideas and tasks
├── debug/ # Active debug sessions
├── milestones/ # Archived milestone data
├── codebase/ # Codebase analysis (from /fh:map-codebase)
└── phases/ # Phase plans, summaries, and verifications
GSD is the state machine. It owns project structure: phases, milestones, requirements, roadmaps, STATE.md. Always required — run /fh:new-project first.
Composites are the interface. They wire engineering disciplines, design quality tools, and GSD together so you never think about which skill to invoke. Subagents do the heavy work — composites stay under 15% context.
Self-contained. All engineering disciplines, design quality tools, GSD project tracking, and TypeScript LSP integration are built into this plugin.
Every composite that executes code enforces these via built-in skills:
/fh:review --quickTypeScript LSP (installed via /fh:setup) provides precise code navigation used across composites: workspaceSymbol and documentSymbol for architecture analysis, findReferences for blast radius mapping, goToDefinition and call hierarchy for data flow tracing, hover for type information.
npx claudepluginhub cinjoff/fhhs-skills --plugin fhEnforces invoking Skill tool before any response or clarification when starting conversations, guiding proper skill discovery and usage workflow.
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.