Menu
Browse Agent Kit documentation directly in terminal. Use for quick access to architecture, skills, workflows, and reference docs.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | agentdocs |
| description | Browse Agent Kit documentation directly in terminal. Use for quick access to architecture, skills, workflows, and reference docs. |
| disable-model-invocation | true |
| allowed-tools | Read, Glob, Bash |
| argument-hint | ["section"] |
Browse the Lucid Labs Agent Kit documentation without leaving the terminal.
/agentdocs # Show documentation overview
/agentdocs [section] # Jump directly to section
/agentdocs search [query] # Search across all docs
When called without argument, show this menu:
┌─────────────────────────────────────────────────────────────────────────────────┐
│ │
│ AGENT KIT DOCUMENTATION │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ GETTING STARTED │
│ [1] overview - Was ist das Agent Kit? │
│ [2] quickstart - Schnellstart für neue Projekte │
│ [3] stack - Tech Stack Übersicht │
│ │
│ ARCHITECTURE │
│ [4] architecture - Systemarchitektur & Patterns │
│ [5] aidd - AIDD Workflow (Adaptive AI Discovery & Delivery) │
│ [6] piv - PIV Loop (Plan-Implement-Validate) │
│ │
│ DEVELOPMENT │
│ [7] skills - Alle verfügbaren Skills │
│ [8] mastra - AI Agents & Tools mit Mastra │
│ [9] convex - Realtime Database │
│ [10] design - Design System & UI Patterns │
│ │
│ INTEGRATION │
│ [11] linear - Linear Project Management │
│ [12] mcp - MCP Server Setup │
│ [13] productive - Productive.io Time Tracking │
│ │
│ DEPLOYMENT │
│ [14] deployment - Docker, Elestio, CI/CD │
│ [15] scaling - Skalierung & Performance │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ Wähle [1-15] oder tippe einen Suchbegriff: │
│ > _ │
│ │
└─────────────────────────────────────────────────────────────────────────────────┘
| Section | File(s) |
|---|---|
overview | .claude/agent-kit-for-ai.md |
quickstart | CLAUDE.md (Quick Start section) |
stack | .claude/agent-kit-stack-overview.md |
architecture | .claude/reference/architecture.md |
aidd | .claude/reference/aidd-methodology.md |
piv | CLAUDE.md (PIV Loop section) |
skills | .claude/skills/README.md + list all skills |
mastra | .claude/reference/mastra-best-practices.md, mastra/README.md |
convex | convex/README.md |
design | .claude/reference/design-system.md |
linear | .claude/reference/linear-setup.md |
mcp | .claude/reference/mcp-servers.md |
productive | .claude/reference/productive-integration.md |
deployment | .claude/reference/deployment-best-practices.md |
scaling | .claude/reference/scaling.md |
/docs overview┌─────────────────────────────────────────────────────────────────────────────────┐
│ │
│ AGENT KIT OVERVIEW │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ Das Agent Kit ist ein modulares Engineering Toolkit für AI-Agent-Projekte. │
│ │
│ KERNKONZEPTE: │
│ │
│ • Upstream/Downstream Model │
│ - Agent Kit = Upstream (Template) │
│ - Dein Projekt = Downstream (Fork) │
│ - /sync holt Updates, /promote teilt Verbesserungen │
│ │
│ • AIDD Workflow │
│ - Exploration → Decision → Delivery → Review │
│ - Linear-basiertes Ticket-Management │
│ │
│ • PIV Loop │
│ - Plan → Implement → Validate │
│ - Keine Phase überspringen! │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ Weiter: [stack] [architecture] [skills] [zurück] │
│ │
└─────────────────────────────────────────────────────────────────────────────────┘
/docs skills┌─────────────────────────────────────────────────────────────────────────────────┐
│ │
│ VERFÜGBARE SKILLS │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ SESSION │
│ /prime Load project context, start session │
│ /session-end End session, update Linear, save time │
│ │
│ PLANNING │
│ /create-prd Create Product Requirements Document │
│ /plan-feature Plan feature implementation │
│ /init-project Initialize new downstream project │
│ │
│ IMPLEMENTATION │
│ /execute Execute an implementation plan │
│ /commit Create formatted git commit │
│ │
│ VALIDATION │
│ /visual-verify UI verification via agent-browser │
│ /pre-production Security & quality check before deploy │
│ │
│ INTEGRATION │
│ /linear Linear project management │
│ /productizer Bridge Linear ↔ Productive.io │
│ /notion-publish Publish markdown to Notion (private) │
│ /clone-skill Clone skills from central repository │
│ /publish-skill Share your skills with the team │
│ /agentdocs This documentation browser │
│ │
│ MAINTENANCE │
│ /sync Sync updates from upstream │
│ /promote Promote patterns to upstream │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ Detail: /docs skill [name] Beispiel: /docs skill commit │
│ │
└─────────────────────────────────────────────────────────────────────────────────┘
/docs stackZeige den Inhalt von .claude/agent-kit-stack-overview.md formatiert an.
/docs skill [name]# Lade Skill-Definition
cat .claude/skills/[name]/SKILL.md
Zeige formatiert:
/docs search [query]# Suche in allen Dokumenten
grep -r -i "[query]" .claude/ --include="*.md" | head -20
Output:
┌─────────────────────────────────────────────────────────────────────────────────┐
│ │
│ SUCHERGEBNISSE: "convex" │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ [1] CLAUDE.md:142 │
│ "Database: Convex - Realtime reactive database" │
│ │
│ [2] .claude/reference/architecture.md:78 │
│ "Convex provides automatic real-time updates to UI" │
│ │
│ [3] .claude/agent-kit-stack-overview.md:45 │
│ "STANDARD: Convex (Realtime, Type-safe, Built-in Vector)" │
│ │
│ ───────────────────────────────────────────────────────────────────────── │
│ │
│ Öffnen: Wähle [1-3] für Details │
│ │
└─────────────────────────────────────────────────────────────────────────────────┘
Innerhalb jeder Section:
| Key | Action |
|---|---|
[zurück] | Zurück zum Hauptmenü |
[weiter] | Nächste verwandte Section |
[1-n] | Springe zu Subsection |
q | Beenden |
# Starte Docs-Browser
/docs
# Direkt zu Skills
/docs skills
# Skill-Details
/docs skill commit
# Suche
/docs search "linear"
# Stack Übersicht
/docs stack
Cross-project audit and sync. Backs up, bootstraps, promotes, syncs, and verifies all downstream projects.
Structured pre-planning research. Explores codebase, asks clarifying questions, and produces a brainstorm output file for /plan-feature input.
Capture knowledge from development sessions. Debug patterns, architecture decisions, framework gotchas, and integration learnings compound over time.
Load project context and show current status. Use at the start of a session or when context is needed.
Run parallel code reviews using specialized agents (security, performance, simplicity, nextjs-react). Produces a structured report.
Initialize a new project from Agent Kit boilerplate. Use when creating a new downstream project.