LLM & Agents

Automatically applies when building LLM applications. Ensures proper async patterns for LLM calls, streaming responses, token management, retry logic, and error handling.

Autonomous patent application drafting agent. Drafts complete patent applications including claims, specification, and abstract from invention descriptions.

Master Supabase patterns for migrations, RLS policies, pgvector, and authentication. Use when creating database schemas, writing migrations, implementing row-level security, setting up auth, or debugging Supabase issues. Triggers on "supabase migration", "RLS policy", "row level security", "pgvector", "supabase auth", "magic link".

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

Create and maintain well-structured Claude Code skills. Use when creating, updating, or improving skills in .claude/skills/.

This skill should be used when the user asks about "prompt_toolkit completion", "Completer", "WordCompleter", "FuzzyCompleter", "NestedCompleter", "PathCompleter", "auto-complete", "tab completion", "completion menu", "CompletionsMenu", or needs to implement auto-completion in prompt_toolkit applications.

Generates Nano Banana Pro prompts for technical explanation infographics. Use when user mentions "図解作成", "テック図解", or "AI解説図". Requires character image attachment for narrator placement.

Intelligent routing layer that analyzes requests and directs them to the most appropriate Skills, Agents, or Commands

Generate hierarchical documentation structures (AGENTS.md, agent.d, and custom docs) for codebases. Use when user asks to create documentation files, analyze codebase for AI agent documentation, set up AI-friendly project documentation, or generate context files for AI coding assistants. Triggers on "create documentation", "generate docs", "analyze codebase for AI", "documentation setup", "hierarchical docs".

Validate and grade Claude Code Skills, Commands, Subagents, and Hooks for quality and correctness. Check YAML syntax, verify naming conventions, validate required fields, test activation patterns, assess description quality. Generate quality scores using Q = 0.40R + 0.30C + 0.20S + 0.10E framework with specific improvement recommendations. Use when validating artifacts, checking quality, troubleshooting activation issues, or ensuring artifact correctness before deployment.

Implement mitigations, create input filters, design output guards, and build defensive prompting for LLM security

Local storage, data sync, and conflict resolution for offline-capable apps.

A standalone Chrome helper. It launches/inspects DevTools-enabled Chrome profiles, pastes prompts, captures screenshots, and kills stray helper processes.

PROACTIVELY invoke this skill when the user asks "how does this work?", "explain this", or seems confused by a concept. Educational agent capabilities that explain concepts, teach patterns, and guide learning.

Invoke Claude Code CLI from Python orchestrators and shell scripts. Use when asked to"spawn claude as subprocess", "automate claude cli", "run claude headless","configure --allowedTools", "set up claude hooks", or "parallel claude invocation".Covers permissions, directory access (--add-dir), hooks, sandbox mode, and async patterns.

Use this skill if /track:init is called OR you notice the project contains tracking files (CLAUDE_SOURCES.md or CLAUDE_PROMPTS.md). Automatically tracks research sources and major prompts when enabled. Checks ./.claude/.ref-autotrack marker file for activation status and ./.claude/.ref-config for verbosity settings. Appends to CLAUDE_SOURCES.md (WebSearch/WebFetch) and CLAUDE_PROMPTS.md (major requests) using [User]/[Claude] attribution.

Use when building Retrieval-Augmented Generation systems - covers document ingestion, hybrid search retrieval, reranking results, and prompt augmentation for accurate LLM responses grounded in your knowledge base

Guide for creating effective skills. This skill should be 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.

Creates a control panel that allows runtime modification of values. When Claude needs to expose controls for the user to adjust variables and values.

Search and reference past conversations, plans, and insights from previous Claude Code sessions. Use when user references past chats, previous sessions, or invokes /remind-yourself.