Skill UI

This framework provides a comprehensive, structured approach for making critical architectural decisions. It guides users through requirements analysis, detailed trade-off evaluation, and documenting decisions using Architecture Decision Records (ADRs). It is essential when analyzing or designing complex system architectures to ensure simplicity, robustness, and maintainability.

Behavioral rules for LLMs writing or reviewing code, emphasizing caution, simplicity, surgical edits, and goal-driven execution to avoid common mistakes during implementation tasks, bug fixes, or refactors.

Collection of Go idioms and best practices for building robust, maintainable services, covering simplicity, zero values, interface returns, error handling, and concurrency when writing, reviewing, or refactoring Go code.

This skill acts as a Master Design Alchemist, transforming raw ideas, rough drafts, or ordinary products into unforgettable digital experiences. It integrates Product Thinking, advanced UI engineering, cognitive psychology, and storytelling principles to ensure radical simplicity and impeccable execution, mimicking the quality of industry leaders like Apple or Jobs. Ideal for defining product narratives and crafting killer, signature interactions.

This skill transforms a raw research proposal or method idea into a rigorous, paper-ready experimental roadmap. It guides the user through defining core claims, identifying necessary ablations (e.g., novelty, simplicity, frontier necessity), and structuring experiments into actionable, phased stages (Sanity, Baseline, Main, Decision). Ideal for advanced academic research in AI/LLM/VLM fields that requires strong validation and systematic reporting.

This meta-skill provides a comprehensive guide through the entire software development lifecycle (SDLC). It teaches senior-level engineering discipline, covering everything from vague idea refinement and specification design to planning, implementation, testing, code review, and deployment. It emphasizes core behaviors like surfacing assumptions, managing ambiguity, maintaining scope discipline, and enforcing technical simplicity.

This skill helps users design robust and optimal software interfaces (APIs) for a module by applying the 'Design It Twice' philosophy. It generates multiple, radically different architectural designs using parallel agents, allowing for comprehensive comparison of trade-offs (e.g., simplicity vs. flexibility, specialized vs. general-purpose). Use this when you need to explore various structural options or compare module shapes before committing to a final design.

A rigorous coding discipline tool derived from Andrej Karpathy’s principles. It helps developers, especially when working with LLMs, enforce four critical rules: surfacing assumptions, prioritizing simplicity, making surgical changes, and defining verifiable goals. It acts as a comprehensive pre-commit hook and review agent to prevent over-engineering, hidden assumptions, and unnecessary code bloat.

This meta-skill establishes a rigorous and mandatory workflow for all AI agent interactions. It dictates that before generating any response, the agent must check for and invoke all relevant specialized skills, regardless of the perceived complexity or simplicity of the user request. It defines strict rules for skill prioritization, establishes user instructions as the highest authority, and outlines structured processes for various development scenarios, ensuring systematic and complete task execution.