LLM & Agents

Use when creating or modifying Claude Code subagents - agent development with frontmatter, tools configuration, and specialization patterns for November 2025 specifications

This skill teaches the agent how to implement the controller layer in the API project. MANDATORY - You MUST read this skill before modifying any controller files.

Tool use patterns for Claude including schema design, tool_choice modes, result handling, parallel execution, error recovery, and extended thinking integration.

Anthropic Claude API development patterns, tool use, Agent SDK, prompt engineering, and best practices.

Meta-skill for creating and managing Claude Code skills. Activated when creating skills, modifying skill rules, or working with the skill system.

Implement caching strategies using @delon/cache. Use this skill when adding memory cache, LocalStorage cache, SessionStorage cache, or cache interceptors for HTTP requests. Supports TTL-based expiration, cache invalidation, cache grouping, and persistent storage. Optimizes performance by reducing redundant API calls and database queries.

Comprehensive code review skill for Wasp applications. Use when reviewing code, PRs, or architectural changes. Covers Wasp-specific patterns, import rules, auth checks, TDD compliance, multi-tenant permissions, and performance. Integrates all CLAUDE.md critical rules.

This skill should be used when moving data between storage locations, changing data structures, renaming fields, or removing deprecated data. Covers schema versioning, safe migration methods, the Foundry unset operator, and idempotent migrations.

Dynamic agent orchestration system that autonomously creates and dispatches specialized sub-agents with pre-selected skill subsets for complex multi-step tasks

This skill should be used when configuring or using the OpenCode CLI for headless LLM automation. Use when the user asks to "configure opencode", "use opencode cli", "set up opencode", "opencode run command", "opencode model selection", "opencode providers", "opencode vertex ai", "opencode mcp servers", "opencode ollama", "opencode local models", "opencode deepseek", "opencode kimi", "opencode mistral", "fallback cli tool", or "headless llm cli". Covers command syntax, provider configuration, Vertex AI setup, MCP servers, local models, cloud providers, and subprocess integration patterns.

Use when integrating OpenAI GPT models for conversational AI. Covers chat completions, streaming responses, function calling, system prompts, token management, cost optimization, error handling, and building specialized agent personas for Physical AI tutoring.

Enhance and refine prompts for AI coding agents using Chain-of-Thought reasoning. Use when user asks to improve, optimize, rewrite, or enhance a prompt. Transforms vague requests into structured, high-context instructions for Claude Code, Codex, or Gemini CLI.

タスクの複雑さに応じて Claude のモデルを自動的に切り替える。複雑なタスク（アーキテクチャ設計、大規模リファクタリング、難解なバグ修正）では opus を、単純なタスク（小さな修正、フォーマット、簡単な質問）では sonnet を自動選択。コストと品質のバランスを最適化。

Local RAG with Ollama and FAISS

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

Systematic technical debt reduction and code optimization agent. Use when refactoring code, reducing complexity, eliminating code smells, improving performance, cleaning up unused code, or modernizing legacy patterns. Handles dependency updates, architecture improvements, and codebase health metrics.

Autonomous development workflow. Generate detailed specs, plans, and tasks for autonomous agent execution with session memory tracking.

Skills for working with the agent-plane project itself. Demonstrates how skills work.

Complete visual QA - screenshots with full-page scroll to trigger animations, then analyze with Claude. Run after CSS/template changes.

Transform messy code into clean code through disciplined refactoring. Code smells, refactoring patterns, and the courage to improve without breaking. Leave every codebase better than you found it.