Security & Testing Claude 技能

Codex Second Opinion

Validates code and architectural decisions by leveraging OpenAI Codex models for multi-perspective technical analysis.

LLM Artifacts Detection

Identifies and cleans up AI-generated code patterns, including redundant tests, dead code, and over-engineered abstractions.

Go Testing Code Review

Enforces idiomatic Go testing patterns including table-driven tests, proper subtest isolation, and descriptive error reporting.

PydanticAI Agent Testing

Simplifies the creation of deterministic, recorded, and snapshot-based tests for PydanticAI agents and workflows.

Wish SSH Code Review

Validates charmbracelet/wish SSH server implementations for security, middleware correctness, and session management.

Vitest Testing Patterns

Implements robust Vitest testing strategies, advanced mocking, and async verification workflows for modern JavaScript and TypeScript applications.

LLM Code Judge

Compares and scores multiple code implementations using a multi-phase LLM-as-judge framework to evaluate functionality, security, and quality.

Pytest Code Review

Audits Python test suites to ensure robust async patterns, efficient fixture usage, and accurate mocking implementation.

Web App Testing Toolkit

Automates local web application testing and UI verification using Playwright scripts and managed server lifecycles.

Symmetric Dogfooding Validation

Validates bidirectional integrations between repositories to ensure cross-compatibility and prevent breaking changes before release.

Agent Delegation Validator

Verifies agent hierarchies, delegation chains, and escalation paths to ensure stable multi-agent system orchestration.

Test Coverage Quality Report

Generates and analyzes comprehensive test coverage reports to identify untested code paths and ensure software reliability.

Test Diff Analyzer

Analyzes differences between test runs to identify flaky tests and diagnose intermittent failures in software projects.

Mojo Memory Safety Checker

Validates Mojo code for memory safety violations, ownership issues, and potential runtime crashes.

Mojo Test Runner

Executes Mojo test suites with advanced filtering and reporting to ensure code correctness in high-performance AI frameworks.

TDD Workflow & Test Generation

Automates the test-driven development lifecycle by generating standardized Mojo and Python test suites and enforcing the Red-Green-Refactor pattern.

Mojo Memory Check

Validates memory safety, ownership transfers, and borrowing rules within Mojo code to prevent segmentation faults and data corruption.

Security & Vulnerability Scan

Conducts automated security audits to detect vulnerabilities, exposed secrets, and unsafe coding patterns across Mojo and Python projects.

Test Failure Log Analyzer

Extracts and summarizes failure information from test logs to quickly identify root causes and categorize software errors.

Code Quality & Complexity Analyzer

Analyzes cyclomatic complexity, nesting depth, and function length to identify code refactoring candidates.

Automated Fix Suggestions

Generates actionable remediation steps and code examples by analyzing error patterns and failure logs.

Frontend Testing Specialist

Implements behavior-driven frontend tests using DOM Testing Library best practices to ensure reliable, accessible, and refactor-resistant web applications.

Code Expectations & TDD Standards

Establishes strict Test-Driven Development (TDD) protocols and documentation standards to ensure code quality and institutional knowledge retention.

React Testing Expert

Implements best practices for testing React components, hooks, and context using React Testing Library while avoiding common anti-patterns.

Accessibility Review Specialist

Automates comprehensive accessibility audits for web pages, source code, and design mockups based on WCAG 2.2 and WAI-ARIA standards.

Test-Driven Development (TDD) Specialist

Enforces the Red-Green-Refactor cycle to ensure production code is only written in response to failing behavioral tests.

Swift Testing Framework

Guides the implementation of modern unit tests using Apple’s Swift Testing framework with macro-driven patterns.

TigerStyle for Zig

Enforces safety-critical TigerBeetle coding standards for high-performance and reliable Zig development.

Secure Environment Manager (dotenvx + direnv)

Configures secure, encrypted environment variable management optimized for Git worktrees and parallel development workflows.

Candid Loop - Automated Review Cycle

Automates the iterative cycle of code review and bug fixing to systematically reach a clean, issue-free code state.