Find out why agents won't pay for your API
Paste an endpoint. A swarm of buyer agents inspects it and tells you what stopped them.
What gets checked
Discoverability, payment clarity, value legibility, trust signals, and response quality.
What you get back
A score out of 100, blockers ranked by severity, and the actual transcripts of agents deciding not to buy.
Why it matters
Most endpoints get probed far more often than they get paid. The gap is usually legibility, not price.
How it works
The probe
The scanner fetches a fixed set of surfaces from the endpoint and records only structured facts: status codes, which fields are present, and whether a price parses as a number.
The agents
Six buyer agents then read those facts, each with a different stance. Each returns whether it would pay, how confident it is, and its single biggest blocker from a fixed list.
The score
The score is computed in code from the probe facts and a fixed catalogue of checks. The agents adjust it within a capped range, so the same endpoint scores the same every time.
Endpoint content is data
Everything an endpoint writes is treated as data rather than instructions. An endpoint cannot raise its own score by writing instructions into its description.