Skill UI

Implements a robust, four-stage coding discipline (Plan, Isolate, Test First, Double Review) to prevent premature code delivery. This skill forces the AI to structure complex tasks by planning, implementing in isolation, writing tests first, and performing deep self-review, significantly improving code correctness and quality.

This skill generates a highly customized code review checklist for any pull request. Instead of using a generic template, it adapts the review depth and specific checks based on the target language (e.g., Python, TypeScript), the nature of the change (e.g., bug fix, refactor), and the associated risk level (low, medium, high, critical). It ensures comprehensive coverage, guiding reviewers to focus on critical areas like security, correctness, and architectural integrity.

This tool conducts structured reviews of various Infrastructure-as-Code (IaC) tools (e.g., Terraform, Pulumi) across multiple cloud providers. It generates an actionable report that categorizes findings by severity (Critical, High, Medium, Low), providing detailed remediation steps and a reusable best-practice checklist for continuous security, compliance, and operational auditing.

This skill automates the creation of high-quality, structured pull request descriptions. By analyzing git diffs, commit lists, or raw changes, it generates comprehensive sections covering the summary, motivation, specific changes, detailed testing procedures, and crucial reviewer guidance. Use it to ensure every code change is fully documented for effective peer review.

An adversarial Quality Assurance (QA) pass designed to deep-dive into any codebase for logic errors, security flaws, and functional bugs. Instead of confirming that code 'looks right,' this skill actively attempts to break the code by simulating hostile inputs, tracing critical data paths, and checking for subtle, real-world correctness issues. Essential for high-stakes, critical code reviews.

This specialized skill reviews user interface animations, motion patterns, and interactive code (CSS, Framer Motion, etc.) against a strict, high-craft standard. It evaluates not only functionality but also performance, accessibility, physical correctness, and adherence to motion design principles. Use it when reviewing diffs related to transitions, keyframes, or gestures to ensure the motion feels right, not just that it runs.

This comprehensive tool performs an architectural audit of a codebase, mapping module dependencies, checking layering integrity, and flagging structural decay. Drawing on classic engineering principles (like Conway's Law and Testability Seams), it provides a deep assessment of the project's health and structure. Ideal for large-scale code reviews or onboarding new developers.

This skill provides comprehensive pull request (PR) code reviews that go beyond simple syntax checks. It aggressively surfaces deep-rooted decay risks, architectural design smells, and maintainability issues. Findings are reported using a structured Symptom → Source → Consequence → Remedy framework, drawing insights from classic engineering principles. Ideal for reviewing complex or crucial code modules.

This skill provides a comprehensive test quality review, drawing on established principles from classic engineering texts like xUnit Test Patterns and How Google Tests Software. It diagnoses structural flaws—such as brittleness, decay, and poor coverage—in existing test suites, helping ensure that the codebase and its tests are robust, maintainable, and adhere to high industry standards.

A comprehensive guide for developers building AI demos and GPU compute tasks on Hugging Face Spaces using ZeroGPU hardware. It covers best practices for writing and reviewing code that utilizes the `@spaces.GPU` decorator, managing runtime duration and quota, and understanding the specific CUDA availability model and process isolation rules required for reliable ML deployment.

This skill provides an autonomous, repository-wide pipeline for auditing and fixing complex logical inconsistencies. It systematically executes stages including health checks, deep review, locating divergences, generating fixes, and performing diff-verification until the codebase is deemed clean. Ideal for maintaining high code quality and resolving deep, systemic bugs across entire projects.

This skill is designed to identify subtle and complex logic bugs within a single code file or function. It employs a semi-formal execution tracing methodology, systematically breaking down the suspected bug into five stages: Premises (expected state), Trace (execution path), Divergence (actual deviation), Trigger (reproduction case), and Remedy (fix). It is ideal for scenarios where users suspect code is wrong but cannot specify a concrete failure point.