LLM & Agents

Search local Claude Code documentation to answer implementation questions. Use when asked about Claude Code features, subagents, workflows, skills, hooks, MCP servers, plugins, settings, CLI options, headless mode, or any Claude Code capability.

Use when generating synthetic training data for multi-turn conversation fine-tuning. Triggers - have design artifacts ready, need to generate conversations, ready to assess quality. Requires finetune-design first.

Expert prompt engineer that transforms vague user requests into optimized, detailed prompts for AI coding agents. Focus is on efficiency - helping users get complete results in one go without follow-up prompts. Triggers on prompt enhancement, prompt optimization, or when refining instructions for AI agents.

Role assignment for Claude Agent

Use when creating or developing, before writing code or implementation plans - refines rough ideas into fully-formed designs through collaborative questioning, alternative exploration, and incremental validation. Don't use during clear 'mechanical' processes

Design reusable prompt templates that encode domain-specific patterns for recurring AI development tasks.Use this skill when you've executed similar prompts 2+ times and need to capture the pattern as reusableintelligence. Transforms one-off prompts into parameterized templates that maintain quality while reducingcognitive load. Helps students move from Layer 2 (AI Collaboration) to Layer 3 (Intelligence Design) inthe 4-Layer Teaching Method.

Auto-invoke when reviewing test files or discussing testing strategy. Enforces testing pyramid, strategic coverage, and stack-appropriate frameworks.

Building LLM agents with LangChain and LangGraph, covering tool-calling model initialization, state management, and observability with LangSmith. Triggers: langchain, langgraph, langsmith, agent-executor, chat-model-tools.

Embedding model configurations and cost calculators

Optimize markdown files for LLM consumption by adding YAML front-matter with metadata and TOC, normalizing heading hierarchy, removing noise and redundancy, converting verbose prose to structured formats, and identifying opportunities for Mermaid diagrams. Use when preparing technical documentation, notes, research, or knowledge base content for use as LLM reference material or in prompts.

Multi-faceted legal operations - generate formal court-ready contracts (senior attorney level) OR simplified explanatory versions. Extends Claude Code beyond coding to professional legal document generation.

This skill should be used when the user asks to "create a workflow", "make a slash command", "add a command", "automate a task with a command", "workflow vs plugin", "should this be a workflow or plugin", or needs guidance on Claude Code command structure, workflow design patterns, or determining appropriate scope for automation requests.

Applies personalized Claude Code settings. Use when setting up Claude Code preferences on a new machine.

Use the hack CLI for local dev environments (docker compose, logs, run commands) and agent integration setup. Trigger when asked to start/stop services, open project URLs, inspect logs, run commands in services, manage branch instances, or configure agent integrations (Cursor/Claude/Codex/MCP). Prefer CLI over MCP when shell access is available.

Invoke this skill first when authoring any AI agent instructions. Foundational principles for writing LLM instructions (skills, CLAUDE.md, rules, commands). Covers token economics, imperative language, formatting for LLM parsing, emphasis modifiers, terminology consistency, and common anti-patterns.

Complete SDK for controlling Reachy Mini robot - head movement, antennas, camera, audio, motion recording/playback.Covers architecture (daemon/client), deployment modes (USB, wireless, simulation, on-Pi), and app distribution.Also includes advanced application patterns: MovementManager, layered motion, audio-reactive movement, face tracking,LLM tool systems, and OpenAI realtime integration.Use when: (1) Writing code to control Reachy Mini, (2) Moving the robot head or antennas, (3) Accessing camera/video,(4) Playing/recording audio, (5) Recording or playing back motions, (6) Looking at points in image or world space,(7) Understanding robot capabilities, (8) Connecting to real or simulated robot, (9) Building conversational AI apps,(10) Integrating with LLMs/OpenAI, (11) Deploying apps to robot, (12) Any robotics task with Reachy Mini.

Write and manage tests using TDD with Pest (PHP), Vitest (React), and Playwright (E2E). Use when writingfeature tests, unit tests, component tests, or improving test coverage. EXCLUSIVE to testing-expert agent.

Designs and configures alerting rules for monitoring systems

This skill should be used when the user asks to "create a Supabase table", "write RLS policies", "set up Supabase Auth", "create Edge Functions", "configure Storage buckets", "use Supabase with Next.js", "migrate API keys", "implement row-level security", "create database functions", "set up SSR auth", or mentions 'Supabase', 'RLS', 'Edge Function', 'Storage bucket', 'anon key', 'service role', 'publishable key', 'secret key'. Automatically triggers when user mentions 'database', 'table', 'SQL', 'migration', 'policy'.

Review test cases for Authorization Endpoint. Covers response_type=code, request parameters (scope, client_id, redirect_uri, state, nonce, prompt, display, max_age), and authorization response per OIDC Core 1.0 Section 3.1.2.