Canva MCP — Workspace Assistant (PoC)
A proof-of-concept MCP server that augments the Canva CLI with custom tools for generating design briefs and staging assets for the Canva API.
README
Canva MCP — Workspace Assistant (PoC)
A proof-of-concept MCP (Model Context Protocol) server that augments the @canva/cli MCP capabilities with custom workspace-assistant tools for design brief generation and asset staging.
Quickstart
# Install dependencies
npm install
# Build TypeScript
npm run build
# Start the custom MCP server (stdio)
npm start
# Development mode with hot-reload
npm run dev
# Launch the native Canva CLI MCP server
npm run mcp
Connecting to MCP Hosts
Claude Desktop / Cursor / VS Code
Copy mcp-config.json into your MCP client configuration and set the environment variables:
| Variable | Description |
|---|---|
CANVA_CLIENT_ID |
Canva Developer App Client ID |
CANVA_CLIENT_SECRET |
Canva Developer App Client Secret |
For Cursor: Settings → Features → MCP Servers → Add new server, paste the mcpServers block from mcp-config.json.
For Claude Desktop: Edit claude_desktop_config.json and add the mcpServers block.
For VS Code: Use the GitHub Copilot Chat MCP configuration or the VS Code MCP extension.
PoC Checklist
This proof of concept validates:
- [x] MCP Server lifecycle — Stdio server start, tool list, and tool call handling via
@modelcontextprotocol/sdk. - [x] Custom tool:
generate_design_brief— Accepts structured input (client name, description, brand colors, deliverables, tone) and returns a formatted Canva Design Kit brief. - [x] Custom tool:
stage_assets— Accepts image URLs and returns a JSON asset manifest that can be consumed by the Canva Assets API. - [ ] Canva SDK compliance — Verify the manifest output matches Canva's
POST /v1/brand-assetsexpected schema. - [ ] End-to-end asset sync — Wire the
stage_assetsmanifest to the Canva CLI MCP for actual upload.
Project Structure
canva-mcp/
├── src/
│ └── index.ts # MCP server entry point, tool registration
├── tools/
│ ├── design-brief.ts # generate_design_brief tool implementation
│ └── asset-stager.ts # stage_assets tool implementation
├── package.json
├── tsconfig.json
├── mcp-config.json # Ready-to-use MCP host configuration
└── .env.example # Environment variable template
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。