web-design-guidelines
Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
vue-layout
Vue 3 + SCSS 切版技能（切版 Agent）— 像素級精確實現。從 Figma 設計稿生成 **1:1 視覺保真的** Vue 3 純展示元件（Dumb Components）。觸發情境包含：「幫我切版」、「根據 Figma 產出元件」、「根據 spec 建立 UI」、「幫我建 Vue 元件」、「切這個畫面」、「實現 Figma 設計」、「補切版」、有提到 Figma URL 或 Node ID 加上 Vue/元件/切版等字詞。保證顏色、間距、字號、圓角、邊框等像素級精確匹配，同時支援與規格流程集成（使用 `draft-spec.md` 或 handoff payload 增強理解）或無規格快速實現。只要上游提供了核准 node 清單，就必須先做 node-to-component mapping 與 coverage 自查，未覆蓋 node 不得宣稱切版完成。僅產出 props/emits 介面的純 UI 結構，不含 API 呼叫或業務邏輯。
logic-coder
Vue 3 前端邏輯技能（邏輯 Agent）。接收 enriched-spec.md（含 API 對應表、State 結構定義、錯誤處理方案），將切版完成的純展示元件（Dumb Components）與 API / 狀態管理邏輯整合，產出可執行的容器層、services、composables。觸發情境包含：接續 api-enrichment 完成後的邏輯實作、「幫我串 API」、「實作資料邏輯」、「加上 loading / error 狀態」、「幫我寫 composable」、「實作 Pinia store」、提到 enriched-spec.md 或 component_paths handoff payload 的情境。
pixel-diff
Pixel-level image comparison between two screenshots. Generates a visual diff overlay highlighting differences. Use for comparing Figma designs against web implementations.
figma-export
Export Figma frames as PNG images via Figma REST API. Use when you need to get a design screenshot from a Figma URL for visual comparison.
figma-implement-design
Translates Figma designs into production-ready application code with 1:1 visual fidelity. Use when implementing UI code from Figma files, when user mentions "implement design", "generate code", "implement component", provides Figma URLs, or asks to build components matching Figma specs. For Figma canvas writes via `use_figma`, use `figma-use`.
api-enrichment
>
ai-pm
>
agent-init
Initialize or improve AGENTS.md files that define how coding agents operate in a repo. Use when asked to set up or replace an agent init command (Codex, Claude), standardize multi-agent behavior, or audit an existing AGENTS.md for clarity, commands, boundaries, and repo-specific context. For Claude Code, also create CLAUDE.md as a symlink to AGENTS.md.
notifications
Set up message notifications (reminders) for plan/triage/review.
google_calendar_local
Read and manage Google Calendar events on this Ubuntu host through a local Python CLI.
interview
Conduct an in-depth interview with the user to create a detailed specification document. Invoke as "/interview [topic]" where topic describes what to interview about. Use when asked to "interview me", "create a spec", "help me spec out", "gather requirements", or when needing to deeply understand a project before implementation.
library-ebooks
>-
llm-seo
Optimize websites and content for AI/LLM discoverability (AIO - AI Optimization). Use when asked to "optimize for AI", "improve AI discoverability", "add LLM SEO", "make site AI-friendly", "help LLMs understand my site", or when implementing llms.txt files, JSON-LD structured data, or AI-focused content strategies.
architecture-review
Review system architecture with an indie hacker lens - bias toward shipping, YAGNI/KISS, fewer moving parts. Invoke as "/architecture-review [path/to/plan]" to roast a plan directory containing overview.md, phases/, and reference/ docs. Use when asked to "review architecture", "roast my design", "check for over-engineering", or "simplify this system".
monta-lights
Control Meross smart plugs (lights). Use when the user says "turn on the lights", "turn off the lights", "lights on", "lights off", or asks about light/plug status.
clean-history
Reimplement the current branch on a new branch with a clean, narrative-quality git commit history suitable for reviewer comprehension. Use when the user wants to clean up messy commit history before opening a PR.
ddia-principles
Apply principles from "Designing Data-Intensive Applications" (Kleppmann) to system design, architecture review, and technology selection. Use when (1) designing new data systems or APIs, (2) choosing between databases, queues, or storage technologies, (3) reviewing architecture for reliability, scalability, or consistency concerns, (4) discussing trade-offs in distributed systems, (5) reviewing code for data handling patterns. Optimized for startup/small team contexts where pragmatic choices matter.
dead-code-detector
|
todo
Create a Plane task in Plane. By default, use the interview skill first to flesh out the issue before creating it; only skip the interview when the user clearly wants a minimal quick capture.
repo-elegance-review
Audit software repositories for elegance/cleanliness/organization and explain why they feel messy or polished. Use when a user asks if a repo is pretty/elegant/clean, wants a codebase aesthetics review, or needs actionable steps to improve structure, naming, cohesion, repo hygiene, or documentation.
remotion-best-practices
Best practices for Remotion - Video creation in React
refine-prototype-to-app
Critique a canonical UI prototype, refine the prototype first, then implement and verify the corresponding application UI against it. Use when a user wants a design critique followed by prototype updates and production parity, especially for iOS or SwiftUI routes with multiple states, sheets, alerts, dialogs, accessibility requirements, or platform-version constraints.
refactor-pass
Perform a refactor pass focused on simplicity after recent changes. Use when the user asks for a refactor/cleanup pass, simplification, or dead-code removal and expects build/tests to verify behavior.
ouraclaw
Fetch Oura Ring sleep data using the ouraclaw CLI. Use when the user asks about their sleep score, sleep data, sleep stages, HRV, heart rate during sleep, bedtimes, or any Oura Ring data. Triggers on "sleep score", "how did I sleep", "oura data", "sleep data", "last night's sleep", "sleep quality", "HRV", or any request for Oura Ring metrics.
triage
Review recent tasks and decide next actions.
un-ai-writing
Rewrite and edit existing text to sound natural and human, removing common AI writing tells while preserving meaning and facts. Use when asked to humanize, de-AI-ify, un-AI, make it sound less robotic, polish an AI draft, or remove AI-writing signs in any prose (emails, docs, posts, essays, scripts).
update-docs
Update README.md, architecture documentation (ARCHITECTURE.md, docs/), and plan files (plan/, task_plan.md, progress.md) to reflect current codebase state. Use when the user explicitly asks to "update docs", "update documentation", "update the readme", "update architecture docs", "update plan", "sync docs with code", or similar requests. Discovers project structure, analyzes code changes, and updates documentation with concise technical descriptions and ASCII diagrams.
human-articles
Write technical articles, blog posts, and long-form content that reads like it was written by a human, not generated by AI. Use this skill when creating articles, tutorials, case studies, essays, or any written content where authenticity matters. Avoids common AI writing tells documented by Wikipedia and other sources.
signal-deck-design
Create or restyle HTML slideshows in the saved Signal Deck visual language: cool plotting paper, disciplined sans/mono typography, flat technical geometry, measurement motifs, and restrained semantic color. Use when the user requests the Signal Deck look or wants a slideshow to feel like a precise measuring instrument. Do not use for generic presentations without this visual direction.
review
Weekly review from Plane tasks.
supernote-upload
Upload PDF and EPUB files to Supernote Cloud. Use this skill when the user wants to send documents to their Supernote device.
plan
Make today's plan from my Plane tasks.
reminders
Internal helper for creating and managing reminders/notifications.
plane-api
Internal helper for creating/listing/updating Plane work items.
taiwan-traditional-chinese
Taiwan Traditional Chinese (zh-Hant-TW) house style: terminology, punctuation, de-AI tone. Judge by the deliverable, not keywords — if your output will contain Chinese, this applies; read it before the first Chinese character, not after. A git, React or JSON task counts once its output is Chinese: commit messages, PR descriptions, ADRs, README, specs, slide decks, code comments, test-case names, zh-TW locale files. Not for ephemeral chat replies, which follow the conversation style rules already in effect. Also for repairing Chinese: mainland wording (數據/組件), mainland software-industry jargon (落地/對齊/閉環), Taiwan terms (元件/資料/函式), AI 味. Default zh-Hant-TW unless 簡體 is asked. Skip when Chinese is the bug not the output (fonts, encoding, full-width spacing, 繁簡轉換) or when translating into another language.
jest-rtl-testing
Use when writing, reviewing, or debugging Jest + React Testing Library tests, before writing test code or when tests fail
css-flexbox-grid
Master Flexbox and CSS Grid layouts for modern responsive design
css-animations
Create performant CSS animations, transitions, and motion design
css-architecture
Implement scalable CSS architecture patterns - BEM, SMACSS, ITCSS, design tokens
css-performance
Optimize CSS performance - critical CSS, code splitting, purging, bundle analysis
css-fundamentals
CSS fundamentals - selectors, specificity, box model, positioning, units
css-modern
Use modern CSS features - custom properties, container queries, :has(), @layer, CSS nesting
css-tailwind
Build with Tailwind CSS utility-first framework - configuration, customization, best practices
css-sass
Use Sass/SCSS for advanced CSS preprocessing with modern @use/@forward syntax
gws-shared
gws CLI: Shared patterns for authentication, global flags, and output formatting.
gws-gmail
Gmail: Send, read, and manage email.
gws-gmail-send
Gmail: Send an email.
gws-calendar-insert
Google Calendar: Create a new event.
Page 1324 of 1708 · 85360 results
