Cluster
llms.txt and MCP Endpoints
The strongest AI-readiness signals are the ones built for machines, not visitors: a technical llms.txt, a .well-known AI file, and a live MCP endpoint. This cluster covers how to read them.
5 questions about llms.txt and mcp endpoints
- Does having llms.txt mean a company is shipping AI?
No. An llms.txt file is a publishing convention, not proof of AI in production, and a standard human-facing one costs nothing to add. Bassethound classifies llms.txt as standard or technical and weighs it against the rest of the stack.
- How do you detect whether a company exposes an MCP endpoint?
You detect an MCP endpoint by probing the conventional paths a remote MCP server uses (often /mcp or /sse) and confirming the protocol: a JSON-RPC 2.0 handshake over Streamable HTTP or a text/event-stream SSE response. A live endpoint is a top-tier signal that a company built for agents.
- Standard vs technical llms.txt: what's the difference?
A standard llms.txt curates content for language models to read. A technical llms.txt exposes machine-facing endpoints for agents to act on. The difference decides whether the file counts as a shipping signal.
- What is a .well-known AI file?
A .well-known AI file is machine-readable metadata served from a site's standardized /.well-known/ directory (the path prefix defined by RFC 8615) that advertises AI capabilities to agents and crawlers. It is one of the strongest signals a site builds for machines, not just humans.
- What is llms.txt and why does it matter?
llms.txt is a markdown file at a site's root that gives language models a curated map of its most important content. Publishing one signals a company is building for AI consumers, not just human browsers.
Related clusters
Frequently asked questions
Why are machine-facing files a strong signal?
A company only publishes a technical llms.txt, a .well-known AI file, or an MCP endpoint to serve automated agents, so their presence points to AI running in production rather than marketing.
Does every AI company have an llms.txt?
No. llms.txt adoption is uneven, so its absence does not mean a company ignores AI. Bassethound weighs it alongside the rest of the stack rather than treating it as decisive.
Sniff a domain.
Run sniff_domain on any site and read its five-layer dossier in one call.