{
  "object": "doc",
  "title": "Which machine-readable files does Cartroute publish?",
  "description": "OpenAPI, MCP server card, API catalog, llms.txt, robots.txt and sitemap.",
  "section": "Resources",
  "url": "https://cart-route.com/docs/machine-readable",
  "updated": "2026-09-23",
  "prev": "https://cart-route.com/docs/guides/recommendations",
  "next": "https://cart-route.com/docs/support",
  "text": "Cartroute publishes an OpenAPI 3.1 description, an MCP server card, an RFC 9727 API catalog, llms.txt and llms-full.txt, robots.txt, a sitemap and security.txt, and every response carries `Link` headers pointing to the API description. An agent that lands on any URL can discover the whole API from there.\n\n## Which files are published?\n\n| File | Format | Purpose |\n\n| [`/openapi.json`](https://cart-route.com/openapi.json) | OpenAPI 3.1 | Every endpoint, parameter, schema and error. Import into code generators and tool builders. |\n\n| [`/.well-known/mcp.json`](https://cart-route.com/.well-known/mcp.json) | JSON | MCP server card: endpoint, transport, protocol versions, authentication, tools. |\n\n| [`/.well-known/api-catalog`](https://cart-route.com/.well-known/api-catalog) | RFC 9727 linkset | Points to the API's description, documentation and status. |\n\n| [`/llms.txt`](https://cart-route.com/llms.txt) | Markdown | A short index for coding agents integrating the API. |\n\n| [`/llms-full.txt`](https://cart-route.com/llms-full.txt) | Markdown | All of this documentation in one file. |\n\n| [`/robots.txt`](https://cart-route.com/robots.txt) | robots.txt | Allows search, user-fetch and training crawlers everywhere except private account pages. |\n\n| [`/sitemap.xml`](https://cart-route.com/sitemap.xml) | Sitemap | Every indexable URL, with last-modified dates. |\n\n| [`/.well-known/security.txt`](https://cart-route.com/.well-known/security.txt) | RFC 9116 | Security contact. |\n\n## Which Link headers are sent?\n\n```\nLink: </openapi.json>; rel=\"service-desc\"; type=\"application/json\"\nLink: </docs>; rel=\"service-doc\"; type=\"text/html\"\nLink: </.well-known/mcp.json>; rel=\"mcp-server\"; type=\"application/json\"\n```\n\n## Is structured data embedded in pages?\n\nYes, as server-rendered JSON-LD: `Organization` and `WebSite` on every page, `Product` with `AggregateOffer` on product pages, `ItemList` on category pages, `Article` on guides, `TechArticle` on docs, and `BreadcrumbList` throughout.\n\n## Which crawlers are allowed?\n\nAll search and user-triggered AI fetchers (OAI-SearchBot, ChatGPT-User, Claude-SearchBot, Claude-User, PerplexityBot, Perplexity-User, Googlebot, Bingbot, Applebot and others) and training crawlers are allowed on every public page. `/admin`, `/dashboard`, `/go/` and account paths are disallowed."
}