llms.txt and Markdown docs
Everything on this docs site exists twice: as the HTML you're reading and as Markdown built for machine consumption. Agents need clean, link-rich text they can index, retrieve, or load into context. This page explains the three machine-readable forms and how they relate.
The corpus
- /llms.txt, the index: A flat map of every page, with a one-line summary and link for each. It follows the llms.txt convention and is small enough to fit in context. Use it as the starting point when an agent retrieves pages on demand.
- /llms-full.txt, the full corpus: Every docs page concatenated as Markdown in one file. Use it for a long-context session, a RAG ingestion job, or a project-level knowledge file that needs the complete corpus.
- The per-page Markdown twin: Every page is also available as raw Markdown, so an agent can retrieve one page without the rest of the corpus.
All three are generated from the same sources as the HTML site and regenerated with every deploy, so they never lag the published docs.
The Copy-page affordance
You rarely need to remember those URLs. Every page across the Guides, API reference, SDK, and knowledge-base sections carries a Copy page menu that hands the machine-readable forms to your tools directly:
- Copy page as Markdown: The current page's Markdown twin, ready to paste into a chat or context file.
- Open llms-full.txt: The full corpus for tasks that need more than one page.
- Copy MCP install command: The one-line command that adds the hosted MCP server to your client's config, so the agent reading the page can also connect the tools needed to act on it.
- Connect to Cursor / Connect to VS Code: One-click MCP setup for two common agentic editors.
The dashboard offers a narrower version of the same idea. On the email onboarding page, the Copy for AI menu copies the current code sample as a ready-to-paste prompt, or opens it directly in Cursor, Claude, or ChatGPT.
The API reference never drifts
The API reference section is generated from the OpenAPI specification that validates the API. Its Markdown twin comes from the same source. When an endpoint changes, the specification, HTML, Markdown, and llms-full.txt regenerate together. An agent reading the corpus therefore reads the API contract directly. Start at the API reference introduction to see how the reference is organized.
Choosing a form
| You want the agent to… | Hand it… |
|---|---|
| Discover what exists and fetch pages on demand | /llms.txt |
| Have the entire docs corpus in context or in a RAG store | /llms-full.txt |
| Understand one topic precisely | That page's Markdown via Copy page |
| Act on Bird as well as read about it | The MCP server; docs alone cannot send an email |
For a guided, end-to-end setup that installs the MCP server, connects the editor, and wires in the corpus, follow AI onboarding.
Next steps
- AI onboarding: install the MCP server, connect the editor, and wire in the corpus.
- MCP server: let your agent act on Bird as well as read about it.
- Agent skills: pre-built workflows that teach coding agents the Bird surface.