LLM & Agents

テストカバレッジ分析とテスト生成ガイダンス。coverage.jsonを解析し、未カバー行/ブランチを特定、優先度付きテスト作成提案を出力。トリガー: "カバレッジ分析", "coverage analyze", "テストカバレッジ向上", "未カバー行"

Build context-efficient, reusable skills using graph-based workflow orchestration with Claude Code subagents. Combines PocketFlow-inspired graph abstraction with multi-model optimization (Haiku for exploration, Sonnet for analysis). Use when building complex workflows, converting external frameworks to Claude skills, or optimizing for cost and context efficiency.

Create, improve, and troubleshoot Claude Code agent skills. Use when asked to make a new skill, write a SKILL.md file, improve skill discoverability, debug why a skill isn't activating, answer questions about skills, or review skill best practices.

Skill for PHP/Laravel backend development following project conventions. Use when creating or editing PHP code, models, services, controllers, tests, or any backend logic. Loads all backend rules from .claude/rules/backend/ and .claude/rules/dataclasses/.

Create Claude Code hooks for event-driven automation and validation

Use when facing 3+ independent failures that can be investigated without shared state or dependencies - dispatches multiple Claude agents to investigate and fix independent problems concurrently

End-to-end software development workflow agent using memex-cli with multiple AI backends. Orchestrates 5 stages: (1) Requirements Analysis (Claude), (2) Functional Design (Claude), (3) UX Design (Gemini), (4) Development Planning (Codex), (5) Implementation & QA (Codex). Use for complete project development from idea to deployment.

Use when setting up git workflows for Claude Code - installs pre-commit hooks, commit templates with Claude attribution, PR templates, branch naming helpers, and Claude-specific gitignore entries

Validates n8n workflow JSON files, tests APIs with real credentials, auto-fixes common issues, and learns new problems. Use when you need to CHECK, VALIDATE, FIX, TEST, PREFLIGHT, or REPAIR n8n workflow JSON files before importing them.

Weekly meal prep and shopping assistant that selects 3 recipes with varied proteins, generates categorized shopping lists, and creates meal prep plans distinguishing Sunday prep from weeknight cooking. Use when planning weekly meals or creating shopping lists.

Build MCP servers for integrating external APIs with AI agents

Supervised fine-tuning using SFTTrainer, instruction formatting, and multi-turn dataset preparation with triggers like sft, instruction tuning, chat templates, sharegpt, alpaca, conversation_extension, and SFTTrainer.

Workflow for parent agents completing child tasks including pulling changes, merging branches, deleting worker branches, and marking tasks complete

Create or update Claude Code skills. Use when: user wants to create a skill, add a new skill, update skill, or mentions SKILL.md. Includes Skill Chaining documentation.

Execute systematic feature development using EPIC methodology (Explore, Research, Plan, Validate, Implement, Review, Iterate). Use when building features, implementing complex tasks, or following structured development workflows. Delegates exploration, research, planning, validation, and review to specialized agents.

Reference for configuring tool permissions when launching Claude Code agents. Use when setting up --allowedTools flags, restricting file access, or configuring agent permissions.

Use after completing file changes - strongest for source code (AST-aware linting, security, tests), lighter support for markdown/config. Dispatches CodeRabbit reviewer subagent. ALWAYS request review before considering work complete.

Fix Flutter layout/constraint issues when embedding WebF (scrollables, unbounded constraints, viewport sizing). Use when the user mentions ListView/ScrollView, “unbounded height”, infinite constraints, RenderBox errors, or embedding WebF in part of the screen.

Collect git info and prepare PR content for reedom-gh:pr-maker agent.

Multi-agent thread orchestration skill for Claude Code