Command Claude Code
A command that enforces test-driven development, or TDD: write failing tests first, add the smallest code that passes them, then improve the code while keeping the tests passing.
Command Claude Code
Command "verify" from znlnzi/claude-config-studio, covering /verify - 验证协议, 验证步骤, 1. 构建验证, 2. 类型检查 and 3. linting.
Hook Claude Code
Runs when a session starts, executing detect-project.sh via bash. From znlnzi/claude-config-studio.
Hook Claude Code
Runs before the context is compacted for auto tool calls, executing pre-compact-save.sh via bash. From znlnzi/claude-config-studio.
Hook Claude Code
Runs when the agent finishes a response, executing check-memory-saved.sh via bash. From znlnzi/claude-config-studio.
Settings file Claude Code
Agent settings declaring 3 hook events (SessionStart, PreCompact, Stop).
Skill Claude CodeCodex
A collection of backend development patterns for APIs, databases, transactions, migrations, caching, and error handling.
Skill Claude CodeCodex
A general guide to programming best practices and coding standards.
Skill Claude CodeCodex
A session-learning tool that records useful lessons from difficult bugs, undocumented library behavior, and better solutions.
Skill Claude CodeCodex
A configuration command for luoshu's AI-powered memory system, including its language-model service, API key, and search models.
Skill Claude CodeCodex
A project setup guide that scans a code project and helps configure its Claude Code files. It asks up to three questions and supports new setups, existing setups, upgrades, and resets.
Skill Claude CodeCodex
A security review checklist for web applications and APIs, covering secrets, input validation, authentication, permissions, data protection, and incident response.
Skill Claude CodeCodex
A project setup assistant that scans a code project and recommends configuration. It supports new projects, existing configuration, upgrades, and resets.
Skill Claude CodeCodex
A test-driven development workflow, where tests are written before the implementation, the smallest working code is added, and the result is then improved. It covers unit tests for small pieces, integration tests for connected parts, and end-to-end tests for complete user journeys.
Skill Claude CodeCodex
A workflow for forming a temporary team of specialised AI agents for a software task. It selects roles such as architecture, development, design, testing, security, or business planning from the available agent files.
Skill Claude CodeCodex
A repeated code-checking process that runs tests and other quality checks during development and before commits.