llms.txt + llms-full.txt
The index files assistants and crawlers look for first.
Be the answer inside ChatGPT, Claude and Cursor.
Most product questions are now asked to an assistant, not a search bar. Orbizz serves your docs in every format those assistants look for, and shows you which of them are reading you.
The week after you sign up: When someone asks an assistant how to do something with your product, the answer comes from your pages — cited, current, and in the version they're on — instead of a guess from a stale training set.
How it works
- Publish in every machine format. Markdown at every URL, llms.txt, llms-full.txt and an MCP server, generated from the same published pages — no export step.
- Measure the invisible readers. Agent analytics show which models fetched which pages, what they searched, and where they got nothing.
- Close the loop. Empty searches become gap reports. Wrong answers become proposed fixes. Your docs get better where agents are actually looking.
What delivers it
Markdown at every URL
Append .md or send Accept: text/markdown.
MCP server
Answers cite the exact page they came from.
Agent analytics
Per-model, per-route, per-page — including 404s and empty searches.
Gap reports
The questions agents asked that your docs didn't answer.
SkillsComing soon
A per-product skill agents can install.
Other audiences: For startups · For API companies
Start writing for both readers.
Free while in beta. Self-hosted, your content, your database.