AI access

Fetch this documentation as raw markdown — per page, as an index, or as one corpus.

This documentation is built to be read by machines as well as people. Every page is available as clean markdown at a stable URL, with no HTML to strip and no JavaScript to execute.

Any page as markdown

Append .md to any documentation URL:

text
https://guidal.ai/docs/services/postgres      → the HTML page
https://guidal.ai/docs/services/postgres.md   → the same page as markdown

Each markdown file carries its own title, summary and canonical source URL, so a page stays attributable after it has been pasted somewhere else.

The index

text
https://guidal.ai/llms.txt

A compact map of the whole site: every page grouped by section, with a one-line summary and a direct markdown URL. Start here — an agent can read this one small file, decide which pages are relevant, and fetch only those.

The whole corpus

text
https://guidal.ai/llms-full.txt

Every page concatenated into a single document. Use it when you want the complete picture in one request and have the context budget for it. For most tasks the index plus two or three pages is a better trade.

Copy for AI

Every page has a Copy for AI button at the top. It copies exactly the bytes served at that page's .md URL — same title, same summary, same source line — so pasting into a chat gives the assistant the identical content it would have fetched.

Using this with an agent

If you have connected your editor over MCP, the agent already operates the platform directly. These URLs are for the other half: understanding Guidal's concepts rather than acting on your account.

A useful pattern is to point the agent at the index once, at the start of a task:

Read https://guidal.ai/llms.txt, then fetch whichever pages you need to deploy this repository as a web service with a Postgres database.

The agent then pulls the two or three pages that matter instead of guessing or loading everything.

Conventions

The index follows the llms.txt convention: an H1 title, a blockquote summary, then ## sections of - [Title](url): summary links. Tools that already understand that convention need no special handling for this site.

All URLs are stable. Pages may be revised, but a URL that exists will keep pointing at the same topic.

What is not here

There is no REST API reference on this site yet — it will be generated from the API specification rather than written by hand, so that it cannot drift from the actual API. It is deliberately not listed in llms.txt until it exists: an index that promises URLs which return nothing is worse than one that stays quiet.