Scoring
What is the agent-readiness score
The 0-100 score shows how ready your site is for AI agents. How it's computed: 5 weighted categories, gradient checks, active categories.
What it is
The agent-readiness score is a number from 0 to 100 showing how legible and accessible your site is to AI agents and AI search. The higher it is, the more easily ChatGPT, Perplexity, Gemini, and YandexGPT find, read, and use your content.
How it’s computed
The score comes from 5 categories, each with its own weight:
| Category | Weight | About |
|---|---|---|
| Content | 30 | content is machine-readable (Schema.org, Markdown, llms.txt) |
| Discoverability | 25 | you can be found (robots.txt, sitemap, link headers) |
| API / Auth / MCP | 25 | agents can interact (MCP, OAuth, ai-agent.json) |
| Bot access | 15 | AI bots are allowed in (AI rules, content signals) |
| Commerce | 5 | agentic payments (x402, ACP, UCP) |
A category score is the weighted average of its checks (each contributes a gradient value 0-1, not just pass/fail). The overall score is the weighted average of the active categories, normalised to 100.
Active categories
- Mandatory — Content, Discoverability, Bot access — always count.
- Optional — API/MCP and Commerce — turn on only when relevant (intent detected, at least one check passed, or you enabled them). If you’re not an API product, a missing MCP doesn’t “penalise” your score.
Result → level
The final score falls into one of 5 levels: from “Invisible” (0-19) to “Agent-Native” (80-100).
Find out yours
Run a free scan to get your score, a per-category breakdown, and a list of what to fix first.