# Agent ready

> Agent readiness means a non-human client can discover and use your content without parsing your layout. Four public scanners grade it and they measure different things. Two checks, Link headers and Markdown negotiation, are response behaviour that no static host can provide.

This site implements agent readiness and links the evidence rather than describing it abstractly.

## Live machine-readable surfaces

| Surface | What it is for |
| --- | --- |
| [/llms.txt](https://emin.de/llms.txt) | Product definition plus every important page as a Markdown link. |
| [/llms-full.txt](https://emin.de/llms-full.txt) | Every article's complete text in one fetch. |
| [/openapi.json](https://emin.de/openapi.json) | Typed schemas and unique operationIds for every operation, errors included. |
| [/auth.md](https://emin.de/auth.md) | Public reads need no credentials; the corpus endpoint uses x402. |
| [/api/articles](https://emin.de/api/articles) | Unauthenticated JSON list of every article. |
| [/mcp](https://emin.de/mcp) | Streamable HTTP MCP server, five read-only tools, no credentials. |
| [/api/v1/corpus](https://emin.de/api/v1/corpus) | Paid bulk access. Unpaid requests return a real x402 challenge. |
| [/.well-known/ai-catalog.json](https://emin.de/.well-known/ai-catalog.json) | ARD manifest listing every machine entry point. |
| [/sitemap.xml](https://emin.de/sitemap.xml) | Every indexable URL with a lastmod date. |

**Basically,** A site claiming to be agent ready should let you fetch the proof, not read a badge.

## Try the Markdown negotiation

```
curl -sI https://emin.de/ | grep -i '^link:'
curl -s -H 'Accept: text/markdown' https://emin.de/ | head -20
curl -sI -H 'Accept: text/markdown' https://emin.de/ | grep -iE 'content-type|vary'
```

## Key answers

- Agent readiness is a set of HTTP behaviours, not a plugin.
- A pure static host can never pass the Link header and Markdown negotiation checks, because both are per-request response behaviour.
- llms.txt entries must be Markdown links. Bare URLs fail the Lighthouse llms-txt audit, and a present-but-invalid file scores worse than no file.
- isitagentready.com grades discovery plumbing; is-agentic.com grades API honesty. Passing one says little about the other.
- Declaring tools or endpoints you have not implemented is the worst available move: an agent reads your metadata as a promise.
- Scanners add checks over time, so a score from last month is not evidence about today.

Full article: [What agent ready actually means](https://emin.de/posts/what-agent-ready-actually-means).
Source code: [github.com/eminogrande/emin-de](https://github.com/eminogrande/emin-de).