Skip to main content

Using ZenPays docs with AI tools

Every page on this site is published as plain Markdown as well as HTML, so a coding agent can read the real source instead of scraping a rendered page.

Copy one page

Every page has a Copy page button at the top right. It copies that page as Markdown, ready to paste into ChatGPT, Claude, Cursor or Copilot Chat.

View as Markdown opens the same content as a plain file, which is the URL to give a tool that fetches its own context.

Fetch a page directly

Append the page's path to /md/ and add .md:

# The HTML page
https://docs.zenpayz.com/docs/rest-api/endpoints/payments/create-payment-intent

# The same page as Markdown
curl https://docs.zenpayz.com/md/rest-api/endpoints/payments/create-payment-intent.md

The Markdown is the page's own source with the site's plumbing removed. Code samples keep their language labels, so a model can tell the cURL example from the TypeScript one.

Point an agent at the whole site

llms.txt is an index of every page, grouped by section, linking to the Markdown rather than the HTML:

curl https://docs.zenpayz.com/llms.txt

Give an agent that one URL and it can find and fetch any page it needs.

Cursor

Settings → FeaturesDocsAdd new doc, and enter:

https://docs.zenpayz.com/llms.txt

Then reference it in chat with @ZenPays.

Claude Code

Point Claude at the index in a prompt, or add it to your project's CLAUDE.md so it is always available:

ZenPays API reference: https://docs.zenpayz.com/llms.txt
Fetch any page as Markdown from https://docs.zenpayz.com/md/<path>.md

Anything that fetches URLs

Give it https://docs.zenpayz.com/llms.txt and it can navigate from there. No key, no account, no rate limit.

Offline copies

For a model with a large context window, or for a reader with no network, the PDF downloads contain the entire documentation set as four volumes. The complete edition is a single file covering every published page.

tip

When you ask an AI to write ZenPays integration code, paste the specific endpoint page rather than describing the endpoint. Amounts in particular are easy to get wrong from memory — every amount is an integer in the currency's smallest unit, and the divisor is not always 100.