Public report · Santos discovery index
jsfiddle.net
Agent-readiness score 69/100 · audited 2026-10-10
Scores
69
Agent-Ready
30
Callable
84
Trustworthy
80
Discoverable
80
Understandable
What this costs you
AI agents can partly use this site, but the gaps below cost citations, recommendations, and automated purchases. These fixes are small compared with the demand they unlock.
Critical exposures
- Machine interfaces not advertised. Nothing in the HTML or HTTP headers points agents to your machine-readable surfaces. If an agent can't find the interface, the interface doesn't exist to it.
- API not described as a service. Your API exists but isn't described with WebAPI metadata. Procurement agents comparing vendors can't classify it, so it drops off shortlists.
Operational friction
- High Integration Tax for AI Procurement Agents: no OpenAPI contract or MCP server. The site takes real inputs, but agents have no typed way to submit them. Each integration needs a human engineer, so autonomous buyers route to vendors who are callable today.
- Pricing not machine-readable. Price, unit, and network aren't published in a form agents can read. An agent that can't quote you can't choose you.
Issues & fixes
- Missing lang attribute on <html>
- Missing Content-Security-Policy header
Embed the badge
Show the live score in your README. The badge updates on every re-audit and greys out when the report is older than 30 days.
[](https://www.santosautomation.com/reports/jsfiddle.net)