Our story
We built Manifest because agents can see the web but can't act on it
Manifest started as an observation: give an AI agent a browser and it can see a page. Give it a scraper and it can read one.
Neither tells it what it's actually allowed to do there what's clickable, what's required, what depends on what.
Every agent framework was solving that problem itself, badly, one hand-written selector at a time.
Manifest is the layer we wished existed: point it at a URL, get back a structured action manifest an agent can reason about directly.
We're Omfang AB, a small Swedish company building infrastructure for the agentic web.
Who we're for
Developers building AI agents
If you're building anything that needs an agent to interact with a real webpage filling forms, navigating flows, completing tasks - Manifest gives you the action layer instead of making you hand-roll selectors that break on the next redesign.
