llms.txt

/llms.txt is a small, human-readable markdown file at the site root that advertises the machine-readable surfaces of an API. It exists to give an LLM crawler or autonomous agent a single entry point that lists every other discovery URL.

Big Balls Sports Data serves one at https://bigballsdata.com/llms.txt.

What it lists

  • API base URL and authentication model
  • Pointers to /openapi.json and /mcp and /.well-known/ai-plugin.json
  • The primary REST endpoints with one-line descriptions
  • Supported sports
  • Tier-by-tier rate limits
  • WebSocket room model and event types
  • Contact addresses for developers and sales

When to read it

For an agent or crawler that does not yet have a tool router configured, /llms.txt is the cheapest first request. It is small, requires no authentication, and tells the caller everything it needs to decide whether and how to proceed.

For an agent that is already aware of the MCP manifest or OpenAPI spec, fetching /llms.txt is redundant, the same information lives in those documents with more structure.

Convention

The file follows the llmstxt.org convention: a top-level H1 with the product name, a blockquote tagline, then H2 sections with bullet lists. The file is intentionally not auto-generated, it stays small and hand-curated.

Content-Type

httphttp
HTTP/1.1 200 OK
Content-Type: text/plain; charset=UTF-8
Cache-Control: public, max-age=0

Served as text/plain per the llmstxt.org convention; the body is markdown so renderers that understand it can format on the fly.