AI assistants

These docs are published for machines as well as people. Every page has a plain-Markdown version, and the whole documentation is available as one file following the llms.txt convention. An assistant that can fetch a URL can integrate Anypost without you pasting docs into the chat.

Machine-readable docs

URL	Contents
`/llms.txt`	An index of every page with a one-line description, per llmstxt.org.
`/llms-full.txt`	The entire documentation as a single Markdown file.
`/docs/<page>.md`	One page as plain Markdown. Append `.md` to any docs path, for example `/docs/send-email.md`.

All three are public, require no API key, and are regenerated from the same source as these pages on every deploy, so they never lag what you are reading.

For most tasks llms-full.txt is the right file: it fits comfortably in a modern model's context window and saves the assistant a page-by-page crawl.

Give your assistant the docs

The fastest path is a single prompt:

Read https://anypost.com/llms-full.txt, then add email sending to this app
using Anypost. Read the API key from an environment variable.

Claude Code

Point sessions at the docs from your project's CLAUDE.md so you never have to repeat yourself:

## Email
 
This app sends email through Anypost.
Full API docs: https://anypost.com/llms-full.txt

Cursor

Add https://anypost.com/llms-full.txt as a custom docs source (@Docs, then Add new doc). Cursor indexes it and pulls relevant sections into context when you reference it.

ChatGPT

Paste the URL into the conversation and ask your question. With browsing enabled, ChatGPT fetches the file and answers from it:

Using the API described at https://anypost.com/llms-full.txt, write a
Python function that sends a templated email with two variables.

Open a page in your assistant

Every page of these docs has a copy menu next to the breadcrumb. It copies the page as Markdown for pasting into any chat, links to the raw .md version, and opens the page directly in Claude or ChatGPT with the Markdown URL preloaded as context.

Where to go next

Quickstart: the four steps your assistant will automate.
Authentication: create the API key the generated code will read.
API conventions: the request and error shapes the whole API shares.