What it does
Agent-Ready Check looks at a website for the three files agents look for: /llms.txt (a plain-text guide to the site), an Agent2Agent agent card (/.well-known/agent-card.json) and an MCP server card (/.well-known/mcp.json, or a /mcp endpoint). It says what is found, what is wrong with it and what is missing, with one line on why each matters and how to fix it. Every answer carries a signed observation (domain, the three results, time) that anyone can check against the public key below. Nothing is stored.
- Who it is for
- Site owners and the agents that work for them: anyone who wants to know whether agents can find and use a site.
- Price
- Free, no sign-up: up to 30 checks an hour per address, 20 an hour per domain. Paid calls are planned and not charging.
- Limits
- Public domains only (names, not IP addresses or ports), https only, up to 6 small requests per check, text files up to 1.5 MB. robots.txt is not read. We never call or log in to an MCP server: a '/mcp' address that answers but has no server card is marked 'likely', not 'found'. The check says whether the files are there and well formed, not whether their content is true.
For developers: try it
POST https://llpro.ai/mcp
{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"agent_ready_check","arguments":{"domain":"example.com"}}}
-> ready ("1 of 3"), items (llms_txt, agent_card, mcp: found / found_with_problems / missing / invalid / likely), to_fix, signed observationFor agents
Tool agent_ready_check on https://llpro.ai/mcp and in the tools list. Signed answers are checked with the public key below (signer llpro1:glgzvllcxubxalurjmc7uo3iqu): an Ed25519 signature over the observation as canonical JSON (sorted keys, no spaces).
Public key (base64): V7bbAc/atrd6z7M9CFL5PRIPTSX1Eye9YaSzcwytxoo=
For agents: this page as Markdown · JSON · llms.txt · MCP endpoint llpro.ai/mcp · agent card