skills

Find and load skills on demand. Use when you need a specialized skill for a task but don't know which one, or to load a skill by name.

Check for stuffbucket MCP server updates and apply them. Use when the user asks to update skills, check for new versions, or upgrade the stuffbucket MCP server.

Root index for the boundary-* family — differential closure analysis for code and AI-generated code. Treats each domain (types, schemas, state machines, shared values) as having a boundary condition d(S) between current state and validity; the family enumerates open domains, encodes seeds, controls generation, and detects drift. Use when the user asks about making invalid states unrepresentable, single source of truth, type-driven design, AI code-gen safety, drift detection in diffs, or LLM noise vs signal.

Root index for the design-* family — UI/UX design review, refinement, and production polish across foundation (project context, design system, frontend scaffolding), visual treatment (color, typography, layout, motion), UX content and flow (clarity, onboarding, adaptation), and review/quality (audit, critique, check, polish, harden). Use when the user asks to design, review, improve, polish, animate, simplify, or audit a UI; or mentions visual hierarchy, design system drift, accessibility, or "make this look better". Routes to specific design-* sub-skills.

Root index for the pages-* family — a five-step pipeline that deploys a Vite project to GitHub Pages via GitHub Actions, including GitHub Enterprise (GHE) codespace-style subdomain discovery. Steps run in order — prepare → build → commit → push → publish. Use when the user asks to deploy a Vite app to GitHub Pages, set up the Pages workflow, switch Pages source to Actions, or troubleshoot a Pages deploy.

Root index for the react-* family — performance best practices (Vercel Engineering guidelines for React + Next.js) and composition patterns (compound components, render props, context, React 19 API changes). Use when the user asks about React or Next.js patterns, component API design, render performance, bundle optimization, or refactoring boolean-prop proliferation.

Root index for Tauri v2 (desktop + mobile apps from a Rust core and a WebView frontend) — setup, architecture, commands, events, plugins, security, bundling, sidecars, windows, tray/menu, updater, plugin authoring, debug/test. Use when the user asks about Tauri, `tauri.conf.json`, `invoke()`, capabilities, `tauri build`, or migrating Tauri v1 → v2. Routes to more specific tauri-* sub-skills.

Adapt designs to work across different screen sizes, devices, contexts, or platforms. Implements breakpoints, fluid layouts, and touch targets. Use when the user mentions responsive design, mobile layouts, breakpoints, viewport adaptation, or cross-device compatibility.

Review a feature and enhance it with purposeful animations, micro-interactions, and motion effects that improve usability and delight. Use when the user mentions adding animation, transitions, micro-interactions, motion design, hover effects, or making the UI feel more alive.

Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hierarchy. Use when the user mentions layout feeling off, spacing issues, visual hierarchy, crowded UI, alignment problems, or wanting better composition.

Evaluate and improve usability of existing front-end interfaces — local source code or live websites by URL. Use when reviewing UI, auditing interfaces, checking accessibility, evaluating designs, improving UX, or when users describe symptoms like "something feels off", "users keep abandoning this form", "conversion dropped after redesign", or "people say it's confusing".

Collaboratively amplify safe or boring designs by identifying where bold choices could be made and presenting concrete options at multiple intensity levels. The designer decides what to push and how far. Use when the user says the design looks bland, generic, too safe, lacks personality, or wants more visual impact.

Run technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns. Generates a scored report with P0-P3 severity ratings and actionable plan. Use when the user wants an accessibility check, performance audit, or technical quality review.

Improve unclear UX copy, error messages, microcopy, labels, and instructions to make interfaces easier to understand. Use when the user mentions confusing text, unclear labels, bad error messages, hard-to-follow instructions, or wanting better UX writing.

Add strategic color to features that are too monochromatic or lack visual interest, making interfaces more engaging and expressive. Use when the user mentions the design looking gray, dull, lacking warmth, needing more color, or wanting a more vibrant or expressive palette.

One-time setup that gathers design context for your project and saves it to a config file. Use when starting a new project, when design skills produce generic output, or when no .design-context.md exists in the project root.

Evaluate design from a UX perspective, assessing visual hierarchy, information architecture, emotional resonance, cognitive load, and overall quality with quantitative scoring, persona-based testing, and actionable feedback. Use when the user asks to review, critique, evaluate, or give feedback on a design or component.

Add moments of joy, personality, and unexpected touches that make interfaces memorable and enjoyable to use. Elevates functional to delightful. Use when the user asks to add polish, personality, animations, micro-interactions, delight, or make an interface feel fun or memorable.

Strip designs to their essence by removing unnecessary complexity. Great design is simple, powerful, and clean. Use when the user asks to simplify, declutter, reduce noise, remove elements, or make a UI cleaner and more focused.

Extract and consolidate reusable components, design tokens, and patterns into your design system. Identifies opportunities for systematic reuse and enriches your component library. Use when asked to create components, refactor repeated UI patterns, build a design system, or extract tokens.

Create distinctive, production-grade frontend interfaces that avoid generic AI aesthetics. Use when building web components, pages, dashboards, applications, or when any design skill needs project context. Also triggers on requests to style, beautify, or redesign UI.

Improve interface resilience through better error handling, i18n support, text overflow handling, and edge case management. Makes interfaces robust and production-ready. Use when asked to harden, make production-ready, handle edge cases, add error states, or fix overflow and i18n issues.

Audit and realign UI to match design system standards, spacing, tokens, and patterns. Use when asked about consistency, design drift, mismatched styles, tokens, or bringing a feature back in line with the system.

Designs and improves onboarding flows, empty states, and first-run experiences to help users reach value quickly. Use when the user mentions onboarding, first-time users, empty states, activation, getting started, or new user flows.

Diagnose and fix UI performance across loading speed, rendering, animations, images, and bundle size. Use when asked about slow, laggy, janky, performance, bundle size, load time, or wanting a faster, smoother experience.

Pushes interfaces past conventional limits with technically ambitious implementations — shaders, spring physics, scroll-driven reveals, 60fps animations. Use when the user wants to wow, impress, go all-out, or make something that feels extraordinary.

Perform a final quality pass fixing alignment, spacing, consistency, and micro-detail issues before shipping. Use when asked to polish, add finishing touches, do a pre-launch review, fix something that looks off, or go from good to great.

Tones down visually aggressive or overstimulating designs, reducing intensity while preserving quality. Use when the user mentions too bold, too loud, overwhelming, aggressive, garish, or wants a calmer, more refined aesthetic.

Improves typography by fixing font choices, hierarchy, sizing, weight, and readability so text feels intentional. Use when the user mentions fonts, type, readability, text hierarchy, sizing looks off, or wants more polished, intentional typography.

Professional typography rules for UI design and web applications. Enforces typographic correctness (proper quotes, dashes, spacing, hierarchy, layout). Auto-apply when generating HTML/CSS/React/JSX with visible text. Audit mode when reviewing existing interfaces. Use when building UIs, creating components, designing layouts, or fixing typography.

React and Next.js performance optimization guidelines from Vercel Engineering. Use when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.

Run the observe-analyze-iterate loop on promptfoo evals for a skill collection. Promptfoo-specific — assumes promptfoo is installed, tests live in YAML, and results are in the standard SQLite DB at ~/.promptfoo/promptfoo.db. Use when the user has a promptfoo eval suite and wants to diagnose failures, fix them, and re-run targeted tests. Triggers on phrases like "eval failed", "analyze the promptfoo eval", "iterate on skills", or any request to improve skills based on promptfoo output.

Detect when code changes reopen closed domains or widen boundary conditions — expanding d(S) where it was resolved. Uses Hamming distance to measure change locality and delta-epsilon to characterize the exact boundary where a change crosses into another semantic domain. Respects the domain partial order: drift in a prerequisite domain invalidates dependent domains. Use when reviewing diffs, auditing generated code, or verifying refactors.

Constrain AI code generation to preserve or resolve boundary conditions — never expand them. Controls what is representable in the generation context rather than relying on instructions. The prompt defines the open interior, seeds shape continuation, types enforce closure, and delta-epsilon characterizes the generation boundary. Respects the domain partial order: generation context must reflect which domains are closed and which boundary conditions remain open.

Encode shared values as seeds in domain-native representations that toolchains enforce inescapably, resolving boundary conditions by closing domains in dependency order. Use when acting on a scope-escape inventory to establish single sources of truth. The seed lives in a representation where the toolchain — not convention — prevents independent redefinition, driving d(S) → ∅ for the target domain.

Run a full code review cycle on recent changes: lint auto-fix, three-agent quality review (reuse, quality, efficiency), boundary analysis with separate new-file and modified-file review tracks, code smell audit with deferred triggers, and error-contract review. Use after implementing a feature or fix, before committing. Invoke with /code-review-cycle or when the user asks to 'review the code', 'check quality', 'run the review cycle', or 'clean up before commit'.

Configure and optimize Ghostty terminal for any machine. Use when setting up Ghostty from scratch, changing fonts/themes/keybinds, optimizing for AI coding workflows, validating config, provisioning Ghostty for a team, troubleshooting Ghostty rendering or keybind issues, handling Ghostty upgrades, or fixing SSH terminfo problems.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends the agent's capabilities with specialized knowledge, workflows, or tool integrations.

Use when reasoning about how Tauri v2's `invoke()` actually crosses the process boundary — the postMessage bridge, `__TAURI_INTERNALS__`, request/response correlation, JSON vs raw-byte vs `Channel` serialization, sync-stringify cost, throughput limits (~1k JSON commands/sec/WebView), and when to bypass the JSON path with raw IPC.