agent-browser
v1.0.0
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.
frontend-design
v1.0.0
Create distinctive, production-grade frontend interfaces and visually strong landing pages with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, applications, landing pages, prototypes, demos, or game UIs (examples include websites, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics, enforces restrained composition, image-led hierarchy, cohesive content structure, and tasteful motion.
http-e2e-testing
v1.0.0
Implement end-to-end testing for web applications using HTTP requests. Use when the task requires API endpoint testing, request/response validation, or integration testing. Triggers: API testing, endpoint validation, HTTP test, integration test, e2e test.
playwright
v1.0.0
Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.
playwright-interactive
v1.0.0
Persistent browser and Electron interaction through `js_repl` for fast iterative UI debugging.
product-design
v1.0.0
Transform natural language feature requirements into standardized YAML definitions. Use when formalizing user stories, creating feature YAML files, modeling product features, or reviewing feature definitions. Triggers: feature design, product spec, requirement modeling, YAML definition.
project-lint
citado
v1.0.0
Runs the real package-local verification commands needed to ensure agent changes are type-safe, buildable, and lint-clean where those checks are actually defined.
project-lint-creator
v1.0.0
Generate a project-specific lint skill (project-lint) for any codebase. Use when setting up automated code quality gates, creating lint/type-check/build verification workflows, or when the user wants to create a project-lint skill for their repository. Triggers: create project-lint, setup lint checks, add code quality gates, generate lint skill, project verification.
reference-ui-debugging
citado
v1.0.0
Use when the user provides one or more reference webpage URLs and wants the frontend to match their layout, spacing, motion, drag-and-drop, hover, pointer feel, or other interaction details. ALWAYS use this skill for reference-driven frontend work such as "参考这个网站实现", "对齐 Google Drive 的交互", "效果不符合预期", "请按这个网页调样式/行为", or whenever a URL is given as the visual or interaction target. This skill requires using
shadcn
v1.0.0
Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI. Provides project context, component docs, and usage examples. Applies when working with shadcn/ui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for "shadcn init", "create an app with --preset", or "switch to --preset".
skill-creator
v1.0.0
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.
skillctl-operator-guide
skillctl
v1.0.2
Use this skill when an agent needs to understand and operate skillctl capabilities end-to-end (discover, install, version, update, publish, and audit).
think-notes
citado
v1.0.0
Records deviations between agent output and user expectations into docs/skills/think.md. ALWAYS trigger this skill when: a user raises a second-round correction, points out a bug or ambiguity that was missed in the previous agent turn, disagrees with an implementation decision, or explicitly says something like "you missed X", "this is wrong", "you didn't handle Y", "I already said Z". The goal is to capture the gap so it can later be refined into a reusable skill or rule — not just silently fix it in one conversation and forget it. Do not skip this skill just because the fix itself is trivial; the value is in the recorded pattern, not the fix.
ui-design-ref
citado
v1.0.0
前端 UI 设计规范参考与决策辅助。当 agent 需要为 Citado 实现或审查 UI 组件时，使用本 skill 核对 token 选择、状态定义、对比度与语义色规则。适用场景包括：新增交互组件、调整颜色/状态样式、选择 Tailwind 工具类、审查 hover/active/selected 实现是否合规。ALWAYS use this skill when implementing or reviewing UI components in packages/fe.
viewer-review
citado
v1.0.0
Review or implement Citado viewer host/runtime changes. Use when a task touches packages/fe viewer code, viewer-core, viewer-pdf, or viewer-image, especially overlays, panels, drag interactions, external stores, or render-engine lifecycles.
webapp-testing
v1.0.0
Browser-based testing and debugging toolkit for web applications using Playwright. Use when verifying frontend behavior, capturing screenshots, inspecting browser logs, or debugging UI issues in local or deployed web apps. Triggers: web app test, UI test, screenshot capture, browser debugging, frontend verification.
yjs-crdt
citado
v1.0.0
Domain-enhanced guidance for building collaborative features with Yjs CRDT, including architecture patterns, provider choices, and operational troubleshooting.
ylapp-local-bootstrap
v1.1.0
Use for: 新电脑拉取 ylapp 后一键完成本地部署与启动。Triggers: 本地部署, 本地启动, 新电脑环境初始化, 启动 ylapp, local bootstrap, run ylapp locally.