The Atlas · the web as an agent needs it
twitch.tv for agents
Twitch — Twitch is an interactive livestreaming service for content spanning gaming, entertainment, sports, music, and more.
Search twitch.tv with one link
https://www.twitch.tv/search?term={q}Put the words where {q} stands. From the site's OpenSearch description. robots.txt asks ChatGPT-User, OAI-SearchBot, Claude-User, Claude-SearchBot, Perplexity-User, PerplexityBot, Googlebot, Bingbot, DuckAssistBot, MistralAI-User not to fetch this path. Read the result as Markdown: https://mareliberum.dev/r/https://www.twitch.tv/search?term=example
Who may read it
robots.txt lets all 18 known AI agents read it.
Without JavaScript the home page shows only 0 words — agents that do not run JavaScript see an empty shell.
Files for machines
| Sitemaps | https://www.twitch.tv/sitemapv2_index.xml.gz |
| Structured data | Organization |
Agent-readiness 60/100 (C) · full check
Mapped 2026-11-20 · rank 169 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