Mermaid is an MCP server that generates and validates Mermaid diagrams and charts directly through AI assistants. It connects large language models to Mermaid rendering workflows, allowing developers, system architects, and technical writers to convert textual requirements or software structures into visual assets. The server supports the full range of Mermaid syntax, alongside styling options such as background colors and custom themes. It validates diagram markup to help AI models self-correct errors during multi-turn generation. Users can export resulting visualizations into multiple formats, including raw Mermaid definitions, SVG, base64 strings, or direct image files saved to disk for local automation. Additionally, it can output remote links like public mermaid.ink URLs for sharing. Operating over standard input/output, Server-Sent Events, or Streamable HTTP transports, this server provides a flexible bridge for integrating visual system architecture, flowcharts, sequence diagrams, and timelines into agentic development environments.
Category: Design, Media & Creative
Tags: charts, diagrams, markdown, mermaid, visualization
claude_desktop_config.json in Claude Desktop). 2. On macOS, add the following to mcpServers: json { "mcpServers": { "mcp-mermaid": { "command": "npx", "args": [ "-y", "mcp-mermaid" ] } } } 3. On Windows, add the following configuration: json { "mcpServers": { "mcp-mermaid": { "command": "cmd", "args": [ "/c", "npx", "-y", "mcp-mermaid" ] } } } 4. Alternatively, install globally using npm install -g mcp-mermaid and run using transports like mcp-mermaid -t sse. 5. Restart your MCP client to load the server.Part of MCP Servers
You can configure Mermaid in your MCP client settings by calling npx -y mcp-mermaid directly. Alternatively, install it globally using npm install -g mcp-mermaid or run the susuperli/mcp-mermaid Docker container over stdio, SSE, or streamable transports.
The Mermaid MCP server enables AI assistants to generate and validate Mermaid syntax for flowcharts, sequence diagrams, and timelines. It supports custom themes and background colors, and exports output to SVG, base64, files on disk, or shareable online URLs.
Mermaid works with desktop applications supporting the Model Context Protocol, including Claude Desktop, VSCode, Cline, and Cherry Studio. It can also connect via HTTP and SSE to web-based AI platforms.
Mermaid supports standard input and output (stdio) for local client integration, Server-Sent Events (SSE) running by default on port 3033, and Streamable HTTP transport running on port 1122 or custom ports.
Yes, Mermaid is open source software released under the MIT License by hustcc on GitHub.