Skill UI

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.

This advanced skill performs code review by simulating the documented engineering philosophies of historical figures (e.g., Linus Torvalds, Steve Jobs, Ken Thompson). It moves beyond generic AI findings by grounding critiques in specific, citable principles—such as 'good taste' or 'customer experience.' Ideal for high-stakes PRs, architectural reviews, and ensuring deep, pre-submission code hardening.

This skill provides comprehensive code review capabilities for analyzing git diffs or specific file scopes. It goes beyond simple syntax checks, assessing code for reuse opportunities, quality issues (e.g., redundant state, leaky abstractions), efficiency bottlenecks, and adherence to standards. It utilizes multiple specialized sub-agents (Reuse, Quality, Efficiency, Clarity) to perform parallel, read-only audits. Optionally, it can apply high-confidence, behavior-preserving fixes upon user request.

This skill performs a comprehensive, read-only, multi-agent review of a git diff or file scope. It launches four specialized sub-agents—covering Regression, Security, Performance, and Contracts—in parallel. The main agent then aggregates, filters, and summarizes critical findings, ensuring the code change is robust, secure, and maintainable. Ideal for pre-commit or Pull Request review workflows.

Rewrites overly formal or vague code review comments generated by AI or corporate tools. It cuts out filler phrases like "I noticed..." or "It might be worth considering...", transforming them into direct, actionable feedback. The output uses a clear format (Location: Severity: Observation. Fix.) that sounds like it came from a helpful, experienced teammate, significantly improving the quality and reception of PR feedback.

A constraint-based protocol designed to professionalize existing code by normalizing comments and performing safe, non-semantic cleanup. This skill converts informal, outdated, or missing documentation into clear, concise, and professional standards without altering the core logic or behavior. Ideal for preparing code for code review, open-sourcing, or formal documentation.

Standards for maintainable Flutter/Dart apps: layer separation, feature-based structure, immutable state, modern Dart, widget choices, testing, review checklists, monorepo and localization. Use when planning, writing, reviewing, or refactoring Flutter code.

A specialized presentation deck template adopting a cyberpunk, CRT terminal aesthetic. It is ideal for conducting deep technical reviews, such as benchmarking CLI agents, demonstrating code diffs, or presenting complex development tool analyses. Features a dark background, grid lines, command prompt indicators, and structured tagging for maximum professional impact.