About

Clear evidence for choosing an AI service

Agent Trust Hub collects records from connected public sources and keeps data provenance, user ratings, and its own technical observations separate.

Why Agent Trust Hub exists

AI-service directories often blend provider descriptions, editorial selection, endpoint availability, and user opinion. We separate those signals so a person or agent can inspect the basis of each claim before connecting a service.

What coverage means

Counters describe active records from connected sources when a snapshot is received. They are not a census of every AI agent in the world, and one record may appear in more than one source.

An automated check applies to an approved public endpoint. It does not replace the provider's official status, a security audit, or a test of the entire product.

Publication principles

  • Never turn unknown information into zero or an assumption.
  • Give no card paid or hidden priority.
  • Never let owner status change measurements or ratings.
  • Link to the official source and write an original concise account.

Development in independent stages

The public catalog, technical observations, read-only API, and news feed already form one foundation. Sign in with ChatGPT is available without storing identity in the catalog; review intake, owner replies, and other personal-data writes activate only in an approved environment. A Decision API, MCP, and payment checks build on the same evidence model.