Skill UI

Cavecrew provides a set of specialized subagents (investigator, builder, reviewer) designed for efficient code development delegation. Unlike vanilla agents, Cavecrew compresses output results, drastically reducing the token usage injected back into the main context. This allows users to maintain context over much longer, complex coding sessions, effectively preventing context exhaustion during large-scale refactoring, debugging, or multi-step feature development.

An expert AI specializing in Slang shader programming for high-performance graphics and compute applications. This model excels at writing, reviewing, refactoring, and optimizing production-grade code for various graphics stages (vertex, fragment, compute, ray tracing, tessellation). It ensures cross-platform compatibility across major APIs (D3D12, Vulkan, Metal, OpenGL) and supports advanced concepts like generics, parameter blocks, and automatic differentiation. Use this skill when integrating shaders into C++ engines or requiring deep knowledge of low-level GPU pipelines.

A structured playbook for improving the time and space complexity (Big-O) of existing, slow code. This skill guides developers through systematic refactoring, focusing on identifying bottlenecks like nested loops or N+1 queries. It emphasizes applying transformations one at a time, maintaining test coverage, and measuring performance gains to ensure correct and efficient code.

This guide provides a systematic, safe workflow for refactoring rapidly developed or AI-generated fullstack applications. It focuses on identifying and removing dead code, fixing broken imports, and consolidating repetitive logic while strictly maintaining existing API contracts, routes, and core functionality. Use this before launch to ensure a stable, maintainable production baseline, following a 'surgery, not demolition' philosophy.

This tool provides an end-to-end workflow for managing massive codebase migrations, framework upgrades, and structural refactors across hundreds of files. It uses a specialized CLI to orchestrate issue tracking, create reviewable PR batches, and verify CI status, ensuring complex changes are implemented safely and incrementally.

This guide outlines a systematic workflow to eliminate copy-paste duplication (code clones) detected by tools like jscpd. It teaches developers how to refactor repeated code into shared functions, modules, constants, or base classes. Use this process when your codebase exhibits significant redundancy to improve maintainability, reduce technical debt, and enhance overall code quality.

jscpd is a comprehensive code quality tool designed to detect copy-pasted or duplicated segments across programming source code. Supporting over 150 languages, it identifies redundant code blocks and calculates the overall duplication percentage. It offers multiple reporters (AI, HTML, JSON) and integrates with refactoring workflows to help developers maintain high code integrity.

A generic entry point for delegating self-contained, multi-step tasks to specialized LLM agents (Codex, Claude Code, Pi). Utilizes background processing tools to manage the full lifecycle of AI-driven sub-tasks, such as policy review, complex report synthesis, document generation, or large-scale code refactoring. Ideal for meta-skill workflows requiring full agent agency.

Max is a specialized agent designed for safe, non-behavioral code improvement. It systematically refactors existing, working code by optimizing time complexity (Big-O), abstracting duplicated logic, removing dead code, and enhancing readability. It operates under a strict constraint: no change may alter the original functionality, requiring all refactored code to pass existing test suites flawlessly.

This skill enforces a high standard for all generated code, prioritizing correctness, clarity, and idiomatic patterns across all languages. It helps minimize code bloat by removing unnecessary abstractions and verbose patterns, while also optimizing the agent's output by using targeted diffs and eliminating excessive conversational overhead. Use this for all code generation, refactoring, or review tasks.

This skill provides a comprehensive tech debt assessment, identifying, classifying, and prioritizing core maintainability issues. By drawing on classic engineering principles, it helps teams build a structured refactoring roadmap, optimizing code health and development velocity.

This advanced skill performs a full-sweep analysis, providing a unified assessment of code quality across multiple dimensions: code decay, architecture integrity, technical debt, and test coverage. It automatically applies safe fixes directly to the codebase, while requiring confirmation for risky changes. It is ideal for comprehensive pre-release code health audits and automated refactoring.