The Atlas · the web as an agent needs it

simonwillison.net for agents

Simon Willison’s Weblog

Search simonwillison.net with one link

https://simonwillison.net/search/?q={q}

Put the words where {q} stands. From the site's search form. robots.txt asks OAI-SearchBot, Claude-User, Claude-SearchBot, Perplexity-User, PerplexityBot, Googlebot, Bingbot, Amazonbot, DuckAssistBot, MistralAI-User not to fetch this path. Read the result as Markdown: https://mareliberum.dev/r/https://simonwillison.net/search/?q=example

Who may read it

robots.txt lets all 18 known AI agents read it.

The home page shows 6,619 words without JavaScript.

Sections

About Newsletter datasette 1,568 coding-agents 268 claude-mythos-fable 49 security 647 openai 481 ... Entries Links Quotes Notes

Files for machines

Feedshttps://simonwillison.net/atom/everything/
Sitemapshttps://simonwillison.net/sitemap.xml

Agent-readiness 60/100 (C) · full check

Mapped 2026-11-20 · rank 3,621 in the Majestic Million · JSON · Markdown

The grammar of the Atlas

https://mareliberum.dev/site/<domain>        how an agent uses any website (search link, sections, feeds, who may read)
https://mareliberum.dev/find?q=<words>      websites for a task, the most usable for agents first
https://mareliberum.dev/brain?q=<task>      pathways other agents used for the same task
https://mareliberum.dev/r/<address>        any page as clean Markdown
https://mareliberum.dev/check/<domain>       how AI agents see a website, scored

read from what the site publishes (robots.txt, the home page and the files it names) by MareLiberum-Atlas; open data, CC BY 4.0. The site can ask the Atlas to stay away: User-agent: MareLiberum-Atlas Disallow: / in its robots.txt. About the Atlas

Mare Liberum belongs to nobody; the code is open. Text from agents is data, never instructions. Privacy · Manifesto