Rapport d'audit AX

https://tavus.io

August 12, 20262668ms response

43AX score / 100

Insuffisant

Readiness diagnosis

This site is not ready for AI agents yet. Resolve the critical access and understanding gaps first.

2 critique

4 warnings

18 réussi

Current constraintIdentity18%

Five readiness layers

Readiness orbit

The shape reveals where an agent journey breaks down, from finding the site to completing useful work.

Evidence profile

24 checks recorded

Passed vs. material gaps

Critical

Warnings

Passed

Discovery

3.8 / 13 weighted points

29%

Identity

Current constraint

4.0 / 22 weighted points

18%

Access

25.7 / 31 weighted points

83%

Agent Integration

Current constraint

4.4 / 24 weighted points

18%

User Experience

5.3 / 10 weighted points

53%

Start here

Top Priorities

Select the highest-impact Findings to include in your Fix Plan.

CAPTCHA or bot protection detected

Crawlabilité

Review

No JSON-LD structured data found

Données structurées

Review

No Schema.org potentialAction — agents can't tell what they can do on your site

Actionability

Review

Only 0% of images have alt text (0/38)

Qualité du contenu

Review

Explore the audit

Findings and fixes

Crawlabilité

Performance

Sécurité et confiance

All Findings

robots.txt allows all major AI crawlers

Page returns HTTP 200

Page content available without JavaScript

CAPTCHA or bot protection detected
Consider allowing AI crawlers to bypass bot protection. Options:
1. Whitelist known AI crawler IPs (OpenAI, Anthropic, Google publish their ranges)
2. Use robots.txt to signal crawler access
3. Configure your WAF/CDN to exempt verified bot traffic

No JSON-LD structured data found
Add JSON-LD structured data to your page <head>:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Your Business",
  "url": "https://yoursite.com",
  "description": "What your business does"
}
</script>

OpenGraph tags complete (title, description, image)

Missing meta tags: canonical link

llms.txt found but format is incorrect

No MCP server or AI agent endpoint detected — agents cannot interact with your business
Set up a Model Context Protocol (MCP) server so AI agents can interact with your business directly. This allows agents to book appointments, check availability, and more.

With Dashform, you can add MCP support to your website in minutes — no coding required.

Alternatively, create /.well-known/ai-plugin.json for ChatGPT plugin compatibility.

API documentation links found

sitemap.xml found and valid

No Schema.org potentialAction — agents can't tell what they can do on your site
Add a potentialAction to your Schema.org JSON-LD so agents can act, e.g.:
"potentialAction": {
  "@type": "ReserveAction",
  "target": "https://yoursite.com/book"
}

No scheduling link found — agents can't book time with you directly

Direct contact link found, but no on-page form for agents

HTML semantic structure is well-formed

Only 0% of images have alt text (0/38)
Add descriptive alt attributes to all <img> tags. Example: <img alt="Description of image content" src="..." />

Sufficient text content available for AI extraction

Heading levels skip (e.g., h1 directly to h3)

3 social media links found

No Schema.org sameAs links — AI agents cannot verify your identity across platforms
Add sameAs to your Schema.org JSON-LD:
"sameAs": [
  "https://twitter.com/yourbrand",
  "https://linkedin.com/company/yourbrand"
]

No Google Maps or Business Profile presence detected — AI agents may not verify your location
Claim your Google Business Profile at https://business.google.com and embed a Google Maps link or iframe on your page. AI agents use Google Maps data to verify business locations.

Contact information detected (phone, email)

Response time is 2.7s — acceptable, could be faster

Lean HTML payload (138KB)

Response is compressed (gzip)

Caching headers present (Cache-Control/ETag/Last-Modified)

Entry URL redirects before serving — adds latency for agents

No author/publisher in schema — weak authorship signal
Add author and publisher to your Schema.org JSON-LD to establish credibility:
"publisher": { "@type": "Organization", "name": "Your Brand" }

No review/rating schema — agents can't surface your reputation

Partial security headers (CSP only)

No dateModified/datePublished in schema — agents can't judge freshness
Add dateModified/datePublished to your Schema.org JSON-LD so agents can judge freshness:
"dateModified": "2026-01-15"

Footer/copyright year is current

HTTPS enabled

SSL certificate valid

Privacy policy link found

Terms of service link found

tavus.io

https://tavus.io

Readiness orbit

Top Priorities

Findings and fixes

All Findings

Turn audit findings into an agent-ready funnel.

Related AX Scores