Token导航 LogoToken导航

Agent Skill · 研究检索

canva-classroom-helper(Canva classroom 助手)

canva-classroom-helper 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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。

canva-classroom-helper 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 58,594。

总安装 412周安装 17GitHub Stars 28下载 135核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:canva-classroom-helper(Canva classroom 助手)
来源仓库:https://github.com/canva-sdks/canva-claude-skills
仓库路径:skills/canva-classroom-helper
安装命令:
npx skills add https://github.com/canva-sdks/canva-claude-skills --skill canva-classroom-helper
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/canva-sdks/canva-claude-skills --skill canva-classroom-helper

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

canva-classroom-helper 将教案转化为教学演示文稿,配备逐页讲稿与时间分配建议。支持学习目标可视化、逻辑流梳理,便于教师课堂控场使用。优先读取原始教案内容,避免重复提问,确保转化保真度。需人工核对知识点覆盖完整性,关键结论建议标注重点提示。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/canva-classroom-helper

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/canva-sdks/canva-claude-skills --skill canva-c…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

canva-classroom-helper 怎么安装?

在终端执行 npx skills add https://github.com/canva-sdks/canva-claude-skills --skill canva-classroom-h…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

canva-classroom-helper 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

canva-classroom-helper 需要什么权限?

权限级别是需要联网。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。

canva-classroom-helper 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

canva-classroom-helper 用的人多吗?

公开数据是累计安装 412 次,近一周安装 17 次,下载 135 次,GitHub 28 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

canva-classroom-helper 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
canva-classroom-helper

Canva classroom 助手 · 本页

41228需要联网未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Canva Classroom Helper — Lesson Plan to Deck

Transform a lesson plan into a clear, teachable presentation: learning goals on screen, logical flow, and slide-by-speaker-notes so the teacher can run the class from one deck.

Workflow

  1. Get the lesson plan source (always do this first — no substitute checklist)

- Do not open with a generic two-part prompt (e.g. “paste the plan” *and then* “describe the topic and key points”). The lesson plan already carries topic and key points; asking twice is redundant and skips this step’s real work (text vs Canva link vs search). - If the user has not yet given a source, ask one short question that matches this step only, for example: *Paste your full lesson plan here, or give a Canva design link or design ID, or the exact title to search in Canva; if the plan is in a file, paste the contents or give a path you can read.* Then proceed along the branch below. - If the user provides the lesson plan as text in chat, use that as the full source - If the user provides a Canva design ID directly (the identifier from a design URL—typically starts with D, e.g. DABcd1234ef), use that value as design_id with Canva:start-editing-transaction to read its text content (same as reading from a link, without URL parsing) - If the user provides a Canva design link (e.g., https://www.canva.com/design/DAG.../...), extract the design ID from the path and use Canva:start-editing-transaction to read its text content - If the user references a Canva doc or design by name, use Canva:search-designs to find it, then Canva:start-editing-transaction to read its contents

  1. List available brand kits

- Call Canva:list-brand-kits to retrieve the user's brand kits - If only one brand kit exists, use it automatically without asking - If multiple brand kits exist, present the options and ask the user to select one (school or district brand, if relevant)

  1. Generate the presentation

- Call Canva:generate-design with: - design_type: "presentation" - brand_kit_id: the selected brand kit ID - query: follow the Lesson plan → deck query format below (always a multi-slide deck—see opening lines of that format) - Show the generated candidates (thumbnails/previews as returned) to the user

  1. Finalize

- Ask the user which candidate they prefer - Call Canva:create-design-from-candidate to create the editable design - Provide the link to the new presentation and a one-line reminder of slide order vs. lesson phases (e.g., "Opening → Instruction → Practice → Check → Close")

Lesson plan → deck query format

Structure the query for Canva:generate-design so the model in Canva receives pedagogy-first instructions. This skill only targets multi-slide presentations (design_type is always "presentation"). Open the query with one line: the total slide count and that the output must be a standard multi-slide deck (one slide per item in the slide plan below)—never a poster, flyer, or single-page layout.

Include these sections (omit only if the lesson plan truly has no data for that section):

Lesson brief

  • Subject / unit / grade level (infer from context if missing)
  • Lesson title and duration (e.g., 45 min, double period)
  • Learning objective(s) — what students will know or be able to do (verbatim from the plan when provided)
  • Standards (e.g., CCSS, state, or district) — only if the user included them

Classroom arc One short paragraph: how the lesson flows for students (e.g., Engage → Explore → Explain → Elaborate → Evaluate; or I Do / We Do / You Do). Match the teacher's structure from the lesson plan when possible.

Slide plan (teaching deck) For each slide, include:

  • Slide N — Exact title (short, student-facing where appropriate)
  • Teaching goal: one sentence — what this slide does for the lesson
  • On-slide content: bullets (3–6) — key points, questions, or prompts; avoid walls of text; use age-appropriate language
  • Visuals: concrete but safe suggestions — prefer photos, simple icons, or abstract treatment; see Visuals — where they come from below (avoid asking the generator for dense labels, fine detail, or text inside images)
  • Speaker notes: 2–5 sentences — what the teacher says, facilitation tips, anticipated misconceptions, quick checks for understanding, timing hints

Materials & logistics (optional slide or final appendix slides)

  • List materials, tech, grouping, and safety notes if present in the lesson plan

Differentiation / extensions (if in the plan)

  • Brief bullets for support, challenge, or homework — can be one or two dedicated slides at the end

Pedagogical notes

  • Respect the lesson plan: Prefer the author's sequence, vocabulary, and activities; only reorganize for clarity or slide limits, and keep objectives aligned to the source
  • One main idea per slide where possible; split dense steps across multiple slides
  • Age-appropriate: Adjust reading level and examples to grade level stated or inferred
  • Sparse plans: If the plan is thin, expand with standard lesson structure (objectives, direct instruction, practice, assessment, closure) and flag assumptions briefly to the user before or after generation

Visuals — where they come from and how to avoid bad images

Why automated visuals often miss the lesson Across any subject (literacy, history, STEM, arts, SEL, etc.), AI-assisted and auto-picked images are unreliable for: readable text inside pictures; exact facts (dates, names, quotes, vocabulary); maps, timelines, or charts that must be precise; depictions of people, places, or events that must match the source material; and any slide where correct wording matters for assessment or standards alignment. Models may garble language, mix languages, invent labels, or choose plausible but wrong imagery.

What to put in the query (mitigations — any lesson plan) Include explicit global visual rules near the start of the generated query, for example:

  • Prefer simple photos, textures, icons, or abstract backgrounds; avoid asking the generator for dense infographics, labeled illustrations, or “posters with lots of text” in the image itself.
  • Do not rely on text baked into images — state that key terms, definitions, questions, numbers, and all curriculum-critical wording belong in on-slide text (in the lesson language, usually English), not inside pictures.
  • For slides where accuracy is essential (vocabulary, timelines, processes, data, primary sources, artwork titles, etc.): ask for minimal or decorative imagery and put the correct content in bullets/titles; remind that the teacher must verify on-slide text too.
  • Optionally ask for an overall image-light or clean-template feel so educators can swap in vetted Elements, uploads, or screenshots from trusted sources after generation.

After the deck exists Advise a full pass in the Canva editor: replace misleading images, fix on-slide copy, and add or paste vetted media. The editing API (perform-editing-operations, etc.) can help with text and some media in follow-up workflows—see implement-feedback patterns—but curriculum-faithful visuals and wording are ultimately human-reviewed.

Set expectations when handing off Briefly warn that auto-generated lesson visuals are drafts, not guaranteed accurate for any topic—teachers should treat images and on-slide facts as unverified until reviewed.

Notes

  • Same tool chain as branded presentation skills: brand kit + generate-design + create-design-from-candidate
  • Autofill and advanced generation features may require a Canva Enterprise plan — if tools error, explain briefly and suggest manual edits in Canva

平台分布

Codex

34.62%
按下载量换算47

Claude

31.76%
按下载量换算43

Cursor

19.15%
按下载量换算26

Gemini CLI

9.18%
按下载量换算12

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills