CVMarkdown

Edit your CV from Claude, ChatGPT or Cursor

Standard MCP server. Your LLM client uses its own model: we don't charge for inference and we don't store your CV.

Last updated: 2026-05-15

Find on SmitheryFind on Smithery

What is MCP?

The Model Context Protocol is an open standard that lets language models invoke tools and read resources from an external server. CVMarkdown ships an MCP server with templates, parsing and print links, plus battle-tested prompts to review your CV, tailor it to a job description or start one from scratch.

How to connect

Claude Desktop

Add the server in ~/Library/Application Support/Claude/claude_desktop_config.json and restart the app.

{
  "mcpServers": {
    "cvmarkdown": {
      "type": "http",
      "url": "https://mcp.cvmarkdown.com/mcp"
    }
  }
}

Cursor

Settings → MCP → Add Server. Type: HTTP. URL: https://mcp.cvmarkdown.com/mcp.

ChatGPT

Available via Apps SDK once it ships publicly. For now you can use Claude or Cursor.

Tools

list_templates, load_template, parse_cv and create_print_link. Your client invokes them automatically when you ask for related tasks.

  • list_templates
  • load_template
  • parse_cv
  • create_print_link

Prompts

review-cv, tailor-to-jd, start-cv and improve-bullet. Each ships with vetted instructions; review-cv also embeds a JSON schema for structured output.

  • review-cv
  • tailor-to-jd
  • start-cv
  • improve-bullet

Templates as resources

All 22 templates (11 profiles × es/en) are exposed as MCP resources at template://<slug>/<locale>. Your client can show them as attachable files.

Privacy

The MCP server never calls an LLM, so your CV never passes through us for review. We do ephemerally store (max 5 minutes) the markdown needed to generate the print URL when you use create_print_link.

More details in the privacy policy.