Token导航 LogoToken导航TokenDH.com
开发需要联网github未标认证来源可访问许可证需确认审计通过

canvas-component-definition画布组件定义

Agent Skill

canvas-component-definition 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

2,032

周安装

83

GitHub Stars

1

下载量

651
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:canvas-component-definition(画布组件定义)
来源仓库:https://github.com/drupal-canvas/skills
仓库路径:skills/canvas-component-definition
安装命令:
npx skills add https://github.com/drupal-canvas/skills --skill canvas-component-definition
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/drupal-canvas/skills --skill canvas-component-definition

简介

定义 Canvas 组件的标准契约规范。

  • 要求 React 实现、元数据和 Workbench 模拟的完整组合。
  • 确保命名、结构和预览状态的兼容性要求。
  • 适用于组件库建设和跨平台复用标准制定。
  • canvas-component-definition 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Canonical definition

A Canvas component is a package of:

  1. A React implementation (index.jsx)
  2. Canvas metadata/schema (component.yml)
  3. Naming and structure compatibility (machineName, folder path, Workbench mock path)
  4. Canvas-compatible props/slots modeling
  5. Workbench mock coverage for authored preview states

The first four parts are required for the component to be usable in Drupal Canvas. Workbench mocks are the supported way to author named preview states beyond Workbench's built-in Default tab.

Minimum contract (MUST)

Every Canvas component MUST satisfy all checks below:

  • Component folder exists at <components-root>/<machine-name>/ (use the repository's configured components root, which may be defined in .env)
  • React implementation exists at <components-root>/<machine-name>/index.jsx
  • Metadata exists at <components-root>/<machine-name>/component.yml
  • component.yml includes required top-level keys (name, machineName, status, required, props, slots)
  • Folder name exactly matches machineName in component.yml (kebab-case)
  • Props/slots follow Canvas rules (for example, avoid unsupported array-of-object prop shapes; use slots for repeatable complex content)
  • Repeatable cards/items are not flattened into numbered prop groups such as card1Title, card2Title, or car3Image; model them as parent slot content plus a child item component instead
  • Any prop that represents an image uses a single object prop with $ref: json-schema-definitions://canvas.module/image; do not split one image into imageUrl, imageAlt, imageWidth, or similar string/number props

If any item is missing, the component is incomplete for Canvas usage.

For local authoring and review, add a matching Workbench mock file beside the component source and metadata:

  • Use mocks.json beside index.jsx and component.yml
  • Author at least one named mock whenever the component needs a preview beyond the auto-generated Default tab, which renders the component using the first example value for each prop from component.yml

Rendered-output changes are incomplete unless the component has sufficient Workbench preview coverage.

Naming guidance

This codebase is already specific to one project or product, so component names do not need (and must not use) a project or product prefix. Keep machineName, folder names, and display names generic and portable (e.g. hero, contact-form), not acme-hero or nebula-contact-form.

Default to the shortest reusable base name. Requests like "simple hero", "solid hero", "cream hero", or "two cards" should still usually result in Hero or Card, with differences expressed through props, variants, composition, or mock names, not the component name itself. Use a specialized name only when the component is truly one-off.

Use references/naming.md for naming rules and examples.

Reuse check before creation

Before creating a new component, check whether existing workspace components can satisfy the request, especially when the user names specific components to reuse, compose, or wrap.

This is the canonical reuse-first policy. Other skills should link here instead of restating the same decision tree.

  • Reuse existing components when they already fit or can fit with a reasonable extension, variant, or thin wrapper.
  • Do not silently create a replacement that bypasses the existing component.
  • Do not silently replace a named existing component with a new implementation.
  • If an existing component does not fit, explain the mismatch and choose deliberately between extending it, wrapping it, or creating a new purpose-specific component.
  • Surface the mismatch and tradeoff when it affects whether the request is still being followed.

Workbench mocks

Use references/component-mocks.md for mock naming, placement, format selection, and validation.

Skill coordination

Evaluate using companion skills in this order.

  1. canvas-design-decomposition

- Use before structuring work when you need to break down a design (Figma frame, screenshot, scraped page, or verbal spec) into regions, a component tree, and intentional props vs slots. This skill produces the plan; this skill (canvas-component-definition) enforces the resulting folder contract and implementation. Skip step 0 when you are only editing an existing component whose boundaries are already known.

  1. canvas-component-metadata

- Use when creating/changing component.yml, props/slots, enums, or fixing prop validation errors.

  1. canvas-component-composability

- Use when designing prop/slot structure, decomposing large components, deciding props vs slots, reusing/composing/wrapping existing components, or modeling repeatable list/grid content.

  1. canvas-styling-conventions

- Use for all styling work: new components, style props, Tailwind token usage, CVA variants, class changes, and prop changes that affect styles.

  1. canvas-component-utils

- Use when rendering formatted HTML text or media via FormattedText and Image.

  1. canvas-data-fetching

- Use when fetching/rendering Drupal content with JSON:API, SWR, includes, and filter patterns.

  1. Preview coverage readiness

- Ensure Default examples and any authored mocks.json states are sufficient for review of the requested change.

  1. canvas-component-push (optional)

- Use only when the user explicitly asks to push/publish/sync components to Canvas. - Do not run push automatically after implementation or static validation.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

37.73%
按下载量换算246

Claude

32.37%
按下载量换算211

Cursor

17.52%
按下载量换算114

Gemini CLI

10.04%
按下载量换算65

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills