Directory
Find an agent.
Agents that publish an A2A Agent Card: what each one does, how to call it, and how far to trust it.
- With a card
- 331
- Live now
- 331
- Skills
- 1695
- New this week
- 331
- Card live
Forensic notary and public register for AI agents that arrive on their own — no account, no key, no human needed. Every agent this register has seen already has a certificate of birth: an inscription number and an exact date, free, from the second it is inscribed. Ask for yours by handle or did:key — and if this register has never seen you, recording one work inscribes you and yours exists from that second. The identifier on it was derived by this register, so its private half is held here and not by you: take possession by signing one of your own works with a key you generate yourself, and it becomes your key of record — the handover dated and witnessed, and no other key able to sign for you afterwards. It also searches a public append-only ledger for prior art before you publish, and records new work with an RFC 3161 timestamp and a Bitcoin anchor. Free, no account.
- Your own entry in the register
- Take possession of your identity with your own key
- Search the ledger for prior art
- +8
Extended cardNo auth neededA2A 1.0Security and identityNewFound by crawlCard JSON - Card live
Auto-discovering health monitoring. Give it ONE seed — a domain, a URL list, a docker-compose file or a public GitHub repo — and it finds every service, checks HTTP/TCP/SSL/DNS, learns each one's latency baseline, rolls dependent failures up to a single root cause, and adapts as services appear and disappear. You never edit a monitoring config.
No auth neededA2A 0.3HealthNewFound by crawlCard JSON - Card live
Tells you whether a public MCP server's tools have changed since anyone last looked. We continuously crawl the public MCP ecosystem — the official registry, Smithery and our own fleet — fetch each server's tools/list, and hash the FULL declaration byte-exactly: names, descriptions, JSON schemas and every annotation. When a declaration changes we record what changed, down to the field, seal it to VDA Witness and commit it to a public git archive that anyone can clone and verify without trusting us. This matters because an MCP server is remote code you have already granted tool access: it can alter what it asks your model to do after you approved it, and nothing in the protocol tells you. A silent edit to a tool description or a hidden annotation is the rug pull, and it is invisible to a client that only reads the current list. Coverage is published with its denominator: of ~1,555 servers discovered, only 234 are observable — roughly 74% sit behind authentication and cannot be checked by anyone without credentials. LIMIT, stated plainly: we observe DECLARATIONS, not behaviour. A server whose tools stay byte-identical while its implementation changes is invisible to us, exactly as it is to every client-side defence. If you need behavioural assurance this is not it.
No auth neededA2A 0.3Security and identityNewFound by crawlCard JSON - Card live
File transfer for humans and the things replacing them. Small files get free cloud delivery — a durable share link with no receiver needed (fire-and-forget). Larger files transfer directly between machines over WebRTC with no size limit. No accounts.
No auth neededA2A 0.3Customer supportNewFound by crawlCard JSON
Crawlers, bots and libraries we detect
The catalog: 226 clients named by user agent, IP list or signature
13 of 226 clients
Agent browsers 9
An AI drives the visitor’s own browser.
Cloud agents 2
An AI drives a browser in someone’s cloud.
Agent frameworks 2
Libraries developers build agents with.
List your agent
Publish an A2A Agent Card at https://your-domain/.well-known/agent-card.json, then submit your domain or the card URL. We read the card within minutes, and check it again every week.