LLM & Agents

Validate design system consistency, check color usage, ensure dark mode coverage, verify Tailwind config sync. Use when checking design consistency, validating color tokens, ensuring dark mode works, auditing component styling, or maintaining the design system.

Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.

Researches and identifies potential customers, leads, and business opportunities for your product or service. Analyzes your offering, finds relevant companies and decision makers, provides contact information, and suggests outreach strategies. Use when looking for leads, researching target customers, identifying decision makers, or planning sales outreach.

Create LangGraph workflows as NestJS applications under apps/langgraph/. Use same webhook pattern as n8n, receive same parameters (taskId, conversationId, userId, provider, model, statusWebhook). Wrap as API agents with request/response transforms. CRITICAL: Status webhook URL must read from environment variables. All endpoints follow A2A protocol.

Creates a permanent backend instance for Claude Imagine. Use this skill when the user wants to set up a new persistent backend server.

Learn Claude Code in 5 hands-on lessons. From basics to building applications. Use when: user wants to learn Claude Code, asks "how do I...", or is new. Triggers: "learn", "teach me", "how does this work", "I'm new", "course".

Set up conda with multiple group storage locations on HiPerGator

Guide for creating valid skills. This skill should be ALWAYS used when users want to create a new skill or update an existing skill that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations. Use it everytime you need to modify anything in ".claude/skills" folder

General-purpose multi-agent orchestration patterns and tooling. Use when coordinating 3+ sub-agents for complex tasks, managing parallel/sequential work streams, synthesizing distributed outputs, or validating cross-agent work quality. Works with any project structure or workflow system.

Essential framework for creating solid Veo 3 prompts. Use when constructing video prompts, validating prompt completeness, or teaching prompt structure. Defines 8 mandatory components (Subject, Setting, Action, Style/Genre, Camera/Composition, Lighting/Mood, Audio, Constraints) that every prompt must include for professional results.

Generate complete, production-ready Claude Code Skills based on business domain and use cases. Use this skill when you want to CREATE NEW SKILLS, BUILD SKILL PACKAGES, or GENERATE AUTOMATION TOOLS for any business domain. Creates self-contained skill packages with SKILL.md, Python scripts, test data, and sample prompts following best practices.

Generates comprehensive unit tests for the selected code. It follows the project's testing conventions, ensures high coverage, and handles edge cases.

Identifies and removes dead, unavailable, or error torrents from Real-Debrid account using the API. Shows torrent status, filters by health, and safely deletes torrents with missing files. Use this when Jellyfin scans are slow, when troubleshooting 404 errors, to clean account storage, or before major library refreshes.

Kumo development assistant for MySQL database management tool. Use when working on Kumo features, understanding architecture, writing tests, or navigating the codebase. Helps with React components, API endpoints, database features, and Electron app development.

Work with the SQLite database layer including FTS5, sqlite-vec, and schema operations. Use for database queries, indexing issues, or storage debugging.

Vector database configuration and setup for pgvector, Chroma, Pinecone, Weaviate, Qdrant, and FAISS with comparison guide and migration helpers

Teach agents to design TUIs in the tuimorphic style (Bagels/Calcure/Claude Code/Droid-inspired).

When working with JSON data in LLM prompts (especially large arrays or tabular data), consider the token-efficient TOON (Token-Oriented Object Notation) format which reduces tokens by 30-70% while maintaining lossless JSON representation and structural validation. Use for reading/writing .toon files, converting JSON↔TOON, or optimizing structured data for LLM consumption with guardrails like [N] counts and {field} headers.

Multi-channel notification dispatcher. Sends trading alerts, reports, and system notifications via Telegram, Slack, Email, and WebSocket. Supports urgency-based routing and custom formatting.

Configure ADK bidi-streaming for real-time multimodal interactions. Use when building live voice/video agents, implementing real-time streaming, configuring LiveRequestQueue, setting up audio/video processing, or when user mentions bidi-streaming, real-time agents, streaming tools, multimodal streaming, or Gemini Live API.