Skill UI

This skill performs a read-only audit to identify missing, weak, or stale test coverage in codebases. It analyzes full repositories or specific scopes like PRs, features, and APIs to recommend concrete test cases. It prioritizes high-risk areas and ensures behavior is properly covered without executing tests or performing security reviews.

This skill performs comprehensive security audits, threat modeling using STRIDE/PASTA, and infrastructure hardening. It supports Red/Blue Team exercises, OWASP compliance checks, code reviews, and incident response planning. Ideal for securing projects, identifying vulnerabilities, and ensuring robust security posture across development and operational environments.

Comprehensive tool for finding and fixing WCAG 2.2 accessibility violations across codebases or live web pages. Operates in two modes: Report mode generates a prioritized, detailed, non-editing audit report (ideal for code reviews), while Fix mode performs an audit-edit-verify loop directly on the target to apply necessary fixes.

Use the GitHub CLI (gh) to systematically inspect, categorize, and resolve review or issue comments on an open pull request. Covers fetching comment threads, planning fixes with user confirmation, applying code changes, and replying to threads as resolved.

This skill allows you to delegate bounded coding tasks to the Google Antigravity CLI (agy) while retaining review and landing responsibility. You write the brief, dispatch the task, wait for completion, review the generated diff against gates, and commit the changes yourself. It requires the agy CLI installed and authenticated, suitable for orchestrators managing code generation workflows.

Secure AI coding agents with permission boundaries, secret protection, code review gates, and safe sandbox configurations for team environments. Covers Claude Code, Cursor, Copilot, and Codex security implementations including CLAUDE.md configuration, command allowlists, filesystem access controls, and pre-commit hooks for secret scanning.

A set of behavioral guidelines derived from Andrej Karpathy's observations to help users writing, reviewing, or refactoring code with Large Language Models (LLMs). It guides developers to avoid overcomplication, make surgical changes, explicitly surface assumptions, and define verifiable success criteria, promoting clean, simple, and robust code development.

This skill provides a comprehensive guide for optimizing Angular applications. It covers performance bottlenecks, bundle size reduction, and rendering efficiency. Use it when writing, reviewing, or refactoring Angular code to ensure optimal speed and load times.

This skill enables non-interactive usage of GitHub Copilot CLI for code reviews, questions, and snippet generation. It requires explicit user consent before sending local files or context to ensure privacy. Use for secondary advice without manual interaction.

This skill enables ultra-granular, line-by-line code analysis to build deep architectural context before vulnerability or bug finding. It applies First Principles and 5 Whys at a micro scale to maintain a stable mental model. Used for security auditing, architecture review, and threat modeling to reduce hallucinations and ensure bottom-up understanding.

This skill performs a comprehensive architecture audit on codebases by mapping module dependencies and checking layering integrity. It identifies structural decay using principles from classic engineering books. The tool generates Mermaid dependency graphs, assesses testability seams, and applies Conway's Law checks to ensure architectural health. It is suitable for developers needing to understand codebase structure, review folder organization, or detect circular imports before refactoring.

AI code reviewer grounded in 12 classic software engineering books. It detects design smells, tight coupling, missing abstractions, stability gaps and data-integrity risks that ordinary linters and AI tools typically miss.