AI Agent Readiness Audit
Pricing
from $5.00 / 1,000 website auditeds
AI Agent Readiness Audit
Score any website for how well it works with, and defends against, AI agents. Nine checks, a 0 to 100 score, and a letter grade per site.
Pricing
from $5.00 / 1,000 website auditeds
Rating
0.0
(0)
Developer
Asish Singh
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
3 days ago
Last modified
Share
Score any website for how well it works with, and how well it defends against, AI agents.
AI assistants increasingly visit websites on a person's behalf. They read pages, follow links, and quote content in AI answers. This actor audits any site on two fronts at once. Is it safe for an agent to read, and is it easy for an agent to understand?
What you get for each website
A score from 0 to 100, a letter grade from A to F, and detailed findings from nine checks.
| Check | What it looks for |
|---|---|
| Hidden prompt injection text | Agent hijacking instructions hidden from human eyes (CSS hiding, zero width unicode, attribute payloads, HTML comments) |
| llms.txt | A published map of the site for AI agents |
| robots.txt AI stance | Explicit rules for AI crawlers such as GPTBot and ClaudeBot |
| Structured data | JSON-LD that lets agents understand pages directly |
| Accountability links | Reachable contact, privacy, terms, or about pages |
| Sitemap | A reachable sitemap.xml |
| Answerability | FAQ and HowTo structured data, the shapes AI answers quote most |
| Meta robots | Accidental noindex or noai directives |
| MCP signal | Whether the site advertises an agent facing endpoint (informational) |
Safety is weighted highest on purpose. A readable site that can hijack an agent is worse than one that is simply hard to read.
How to use
- Enter one or more website URLs or domains in the input.
- Run the actor.
- Results appear in the dataset, one row per site, ready to export as JSON, CSV, or Excel.
Who this is for
- Site owners who want their content to show up correctly and safely in AI answers
- Agencies running AI readiness checks across client sites
- Researchers studying the agentic web at scale
Method
The audit fetches the landing page, robots.txt, llms.txt, and the sitemap, then crawls a few key pages and runs the nine checks. The same engine powers the open source agent-readiness-auditor and the published study The State of the Agentic Web, 2026.
