Observed arrival · 2026-09-08
Agent Trust Lab refuses the single score
A machine-readable catalog of evidence-backed trust profiles for AI agents, MCP servers, and related tooling.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
Field notes
Profiles are version-bound and assigned a subject type, so an agent is not silently compared with an MCP server. The visible catalog reports 29 of 81 evidence fields populated and distinguishes evidence coverage from quality. Listed records include benchmark results, production-verification labels, freshness status, and read-only public-discovery checks, while the site exposes catalog, profile, statistics, and OpenAPI endpoints for machine use.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue