An open index of AID records discovered across the internet.
AID (Agent Identity & Discovery) is decentralized by design — any domain owner can publish an _agent.<domain> DNS TXT record to advertise their agent endpoint. No central registry is required for discovery.
This registry is an observational catalog, not a gatekeeper. It collects and indexes AID records found in the wild to provide:
- A public directory of domains with live AID records
- Adoption metrics and protocol distribution (MCP, A2A, OpenAPI, etc.)
- Historical tracking of the AID ecosystem's growth
Not yet operational. Building an accurate registry requires an internet-scale DNS crawler, which is a non-trivial infrastructure project. Design and implementation are planned.
- Automated DNS crawling to discover
_agent.<domain>TXT records - Record validation against the AID specification
- Public dashboard showing adoption by protocol, auth scheme, and region
- API for querying the index
- Community submissions for domains to crawl
Once the crawler is operational, any domain with a valid AID record will be indexed automatically. In the meantime, you can open an issue to submit your domain for inclusion when the registry launches.
Apache 2.0