Desenvolvimento

Use when the user asks for root-cause-first debugging of bugs, test failures, flaky behavior, build issues, or integration defects with tracked verification and regression lock-down.

Use when building Python 3.11+ applications requiring type safety, async programming, or robust error handling. Generates type-annotated Python code, configures mypy in strict mode, writes pytest test suites with fixtures and mocking, and validates code with black and ruff. Invoke for type hints, async/await patterns, dataclasses, dependency injection, logging configuration, and structured error h

Use when Claude Code needs to diagnose or fix a bug against active SpecDD specs.

Reference guide for npm, pnpm, yarn, and bun package managers. Covers workspace configuration, security audits, troubleshooting, and migration between package managers. Use for complex monorepo setup, debugging package issues, or deep package manager reference.

Create structured QA round documents for manual testing phases. Generates tracking docs with issue logging, session planning, and verification workflows.

GitHub issue 本文の `## 受け入れ条件` セクションを抽出し、各項目を自動検査して結果を ✓ / ✗ / ? で報告する read-only skill。`issue-implement` cycle 内で実装・commit 後・cross-review 前に呼び出して受け入れ条件確認を標準化する。

Analisa repositórios GitHub, ferramentas ou frameworks de IA para extrair suas melhores capacidades em habilidades nativas do OpenClaw, ideal para replicar funcionalidades ou criar novas skills a partir de projetos externos.

Uma investigação de causa raiz em 4 fases para qualquer problema técnico, como bugs, falhas de teste ou comportamento inesperado. Ela exige a identificação da causa raiz antes de tentar correções, não permitindo soluções aleatórias.

Run code review against design documents, checking acceptance criteria, security, performance, and test coverage. Use when reviewing implementation quality or running /review.

Guidelines for creating and updating Claude Code skills. TRIGGER when: writing a new SKILL.md, modifying an existing skill, or the user asks about skill conventions.

Coordinator that executes a plan in parallel groups via background subagents. Spawns implementers, then dedicated test-runner / static-analyse subagents, then routes failures back to the original implementer (SendMessage) or a fresh fixer. Fully autonomous — no human-in-loop.

Use when Claude Code needs to refactor code, tests, docs, specs, or project structure while preserving active SpecDD behavior.