# Guides, readiness scores, and what agents actually check

Your **readiness score** is a structured snapshot of how well your origin exposes agent-friendly discovery—not a vanity metric.

## Start with the guides

Our evergreen docs are the fastest path to a passing profile:

| Topic | Guide |
|-------|--------|
| Site map for LLMs | [What is llms.txt?](/guides/llms-txt) |
| Tool exposure | [MCP server cards](/guides/mcp-server-card) |
| B2B product checklist | [Agent-ready SaaS](/guides/agent-ready-saas) |

Guides are stable reference material. When protocols change, we update the doc—not a scatter of old blog posts.

## What the scanner looks for

At submission we probe paths such as `/llms.txt`, `/.well-known/ard.json`, and `/.well-known/mcp/server-card.json`, plus content negotiation where relevant. Site type (SaaS vs content) changes which checks apply.

You can also run external tools—we link **isitagentready** from admin and listing views for a second opinion.

## Commerce and UCP

If you ship with [Universal Commerce Protocol](https://ucp.dev) manifests, filter the public directory with `?ucp=1` and read [COMMERCE-PROTOCOLS.md](/COMMERCE-PROTOCOLS.md) for how we surface commerce-ready listings to agents.

## After you ship specs

[Submit your site](/submit) for curator review. Approved listings appear in the directory and [Agent Tool Finder](/agents/tool-finder); discovered sites can be [claimed](/badge) by verified domain owners.