# For AI agents

> How agents can read alaskanlabs.com. llms.txt and llms-full.txt, a Markdown twin of every page, the templates and Alaskans as JSON, and the crawler policy.

Canonical: https://alaskanlabs.com/docs/for-ai-agents

This site is written to be read by agents as well as people. Everything on it is static, so a plain HTTP GET is all you need.

## Start here

- [`/llms.txt`](/llms.txt): a map of the site in the [llms.txt](https://llmstxt.org) format, with a line on every page and a link to its Markdown.
- [`/llms-full.txt`](/llms-full.txt): the docs, the Alaskans and the templates in one Markdown file.

## Every page as Markdown

Add `.md` to the address of any docs page, post or marketplace page:

| Page | Markdown |
|---|---|
| [/docs/decide](/docs/decide) | [/docs/decide.md](/docs/decide.md) |
| [/alaskans/research-brief](/alaskans/research-brief) | [/alaskans/research-brief.md](/alaskans/research-brief.md) |
| [/workflows/invoice-chase](/workflows/invoice-chase) | [/workflows/invoice-chase.md](/workflows/invoice-chase.md) |
| [/bots](/bots) | [/bots.md](/bots.md) |
| [/blog](/blog) | [/blog.md](/blog.md) |
| [/](/) | [/index.md](/index.md) |

Each HTML page links its twin with `<link rel="alternate" type="text/markdown">`, and each twin names its canonical page on its first lines and in a `Link: rel="canonical"` header.

## Data as JSON

- [`/alaskans.json`](/alaskans.json): every Alaskan's listing (code, name, tagline, what it does, when it runs, its add link).
- `/teams/<slug>.json` and `/workflows/<slug>.json`: each template whole, MIT-licensed, ready to install or fork. For example [/teams/starter.json](/teams/starter.json).

These send `Access-Control-Allow-Origin: *`.

## Structured data

Every page carries JSON-LD: the organisation and site, `SoftwareApplication` for Alaskan, each Alaskan and each Alaskan bot, `ItemList` for the marketplace, `SoftwareSourceCode` for templates, `TechArticle` for docs, `BlogPosting` for posts, and `BreadcrumbList` everywhere below the home page.

## Crawling

[`/robots.txt`](/robots.txt) allows every crawler, AI crawlers included, and lists the [sitemap](/sitemap.xml). Please cite pages by their canonical address on alaskanlabs.com.

## Adding an Alaskan for someone

If a person asks you to add an Alaskan, give them its link, `https://alaskan.ai/b/<code>`, which opens the iPhone app with the code ready, or its 4-digit code to send to their own Alaskan. Codes are in [/alaskans.json](/alaskans.json).