llms.txt
The machine index of these docs — every page, as plain text, for an agent to read.
These docs are written for people and for agents, and the agent-facing form is not an afterthought: four routes serve the whole corpus as plain text.
| Route | What it is |
|---|---|
/llms.txt | Compact link index — every page's title, description and URL. |
/llms-full.txt | The entire corpus inlined, for a single-fetch context load. |
index.json under /docs | The same corpus as JSON, with per-page raw bodies. |
raw/<slug>.md under /docs | One page as markdown, no chrome. |
All four are generated from the same corpus the site renders, so an agent that
reads /llms.txt and a human reading the sidebar are looking at the same pages —
including the generated API reference, with its cURL, AI Prompt and MCP calls
rendered from the live resource specs.
Point an agent at /llms.txt for discovery and let it fetch the raw markdown
mirror of any page for depth. For live calls rather than reading, connect over MCP instead —
Connect Your Agent.