Informe de auditoría AX

https://unravelcarbon.com

August 12, 20261635ms response

61AX score / 100

Needs work

Readiness diagnosis

This site needs work before agents can use it reliably. Address the priority findings to improve readiness.

0 crítico

4 warnings

24 aprobado

Current constraintAgent Integration24%

Five readiness layers

Readiness orbit

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

Evidence profile

28 checks recorded

Passed vs. material gaps

Critical

Warnings

Passed

Discovery

4.6 / 13 weighted points

36%

Identity

16.2 / 22 weighted points

74%

Access

29 / 31 weighted points

94%

Agent Integration

Current constraint

5.7 / 24 weighted points

24%

User Experience

5.3 / 10 weighted points

53%

Start here

Top Priorities

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

No llms.txt file found — AI agents cannot discover your services

Interacción con el agente

Review

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

Actionability

Review

Only 14% of images have alt text (10/72)

Calidad del contenido

Review

Explore the audit

Findings and fixes

Interacción con el agente

Actionability

All Findings

robots.txt allows all major AI crawlers

Page returns HTTP 200

Page content available without JavaScript

No CAPTCHA or bot protection detected

Found 1 JSON-LD structured data block

Business schema found: Organization

Schema has 7 fields — consider adding more for better AI comprehension

OpenGraph tags complete (title, description, image)

Meta tags complete (title, description, canonical)

No llms.txt file found — AI agents cannot discover your services
Create a /llms.txt file at your site root:

# Your Business Name

> Brief description of your business.

## Services
- [Service 1](/service-1): Description
- [Service 2](/service-2): Description

See https://llmstxt.org for the full spec.

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.

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

Contact/lead form with an email field is present

CTA buttons have descriptive labels

HTML semantic structure is well-formed

Only 14% of images have alt text (10/72)
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)

1 social media link found

Schema.org sameAs links found (cross-platform identity)

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 1.6s — acceptable, could be faster

Lean HTML payload (71KB)

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

Security headers present (HSTS + Content-Security-Policy)

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

unravelcarbon.com

https://unravelcarbon.com

Readiness orbit

Top Priorities

Findings and fixes

All Findings

Turn audit findings into an agent-ready funnel.

Related AX Scores