cowart
v0.1.28by zhongerxin · author: ZHONG XIN
A native tldraw canvas for visual thinking and image workflows. Use cowart_mcp to open the canvas when requested and reuse an existing widget for subsequent Cowart work.
What cowart does, in the publisher's words
Cowart 是一个面向 Codex 的原生无限画布 widget 插件。它基于 tldraw 提供可视化画布，用于构思、标注、生成图片和根据标注图迭代图片。画布由 MCP widget 直接打开，数据默认保存到当前用户项目的 canvas/ 目录，而不是保存到插件仓库里。
仓库同时遵循 Agent Plugins v1.0.0：根目录的 plugin.json、skills/ 和 mcp.json 提供可移植插件入口；.codex-plugin/plugin.json、.mcp.json 和.agents/plugins/marketplace.json 保留 Codex 专用的界面与安装元数据。
English README: README.en.md
- 在 Codex 中打开一个原生 tldraw 无限画布 widget；正常使用不再通过网页浏览器或 in-app browser 打开本地页面。
- 在当前项目目录中持久化画布页面和图片资源。
- 在画布中创建 AI 图片框，直接输入 prompt、选择参考图，并让 Codex 按选中框的位置和比例生成图片后替换它。
- 创建 16:9 的 AI HTML 框，通过 prompt 和参考图生成可运行的单文件 HTML，并直接嵌入画布继续编辑或迭代。
- 创建 AI Slides，将图片和 HTML 组织成演示文稿，或让 Codex 按指定页数生成一组 16:9 HTML 页面；支持缩略图预览和全屏播放。
- 标注好图片后，可从画布里直接提交标注截图，让 Codex 根据标注生成干净的新图并放到原图旁边。
From the project README, punctuation lightly normalized · Open the README on GitHub
- Skills
- 3
- MCP servers
- 1
- Stars
- 5,963
- License
- MIT
- Repo created
- 2026-06-18
- Last pushed
- 2026-09-13
- Publisher type
- User
- Version
- 0.1.28
Links
Skills · 3
- cowart-image-edit
- Generate revised AI images from user-supplied Cowart annotation screenshots and place each result beside its original without replacing, moving, hiding, or deleting existing images or annotations.
- cowart-image-gen
- Generate and place a final AI bitmap on the Cowart canvas, replacing a selected AI 图片 holder or inserting the image into the current page.
- cowart-open-canvas
- Open, reopen, or explicitly refresh the native Cowart canvas when the user asks to see it, or when a bare @Cowart invocation has no other requested workflow.
Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.
MCP servers · 1
- cowart_mcpstdio
- node ./scripts/start-mcp.mjs
Read from the plugin's own mcp.json. We list environment variable names, never values.
Category
Creative & Media. Assigned by this directory. Agent Plugins 1.0.0 has no category field, so no manifest declares one.
Keywords
canvas · tldraw · whiteboard · image-generation · visual-collaboration
Related by publisher and keywords
Social media creative pipeline: 20 skills and 5 agents for platform-ready posts, carousels, image/video assets, and compliance-gated delivery.
Generic toolkit for driving a local ComfyUI 8188 server: workflow submission, queue monitoring, image download, and reproducible generation patterns. Includes a dependency-free stdio MCP server and four Skills (comfyui-studio + comfyui-workflow + comfyui-character + comfyui-drama) covering 6 preset scenarios (生图 / 模仿 / 改图 / 融合 / 首帧 / 出片).
Capability-aware MiniMax Token Plan and multimodal API tools for search, vision, images, speech, video, lyrics, music, and covers.
把一段场景描述变成粉红色章鱼打工人的 GIF 表情包：输出 720x720 循环 GIF 和适合聊天发送的 480x480 小图。需要 MiniMax Code 提供 image_synthesize 与 gen_videos 两个宿主工具。 / Turn a scene description into a pink octopus office-worker GIF meme: a 720x720 looping GIF plus a 480x480 mini-GIF for chat apps. Requires MiniMax Code to provide the image_synthesize and gen_videos host tools.
Jon Gallant's collection of cross-agent skills for AI coding agents, GitHub Copilot, Claude, and Codex. Browse and install individual skills from the skills/ directory.
Let AI agents capture live website screenshots, render HTML/CSS, and populate reusable templates as images or PDFs, without managing a browser.
[](https://agentpluginsdirectory.com/plugins/cowart)✓ Verified . We fetched the manifest from GitHub and checked it against the official Agent Plugins 1.0.0 schema at agent-plugins.org.