AX監査レポート

https://loft.sh

August 12, 20263177ms response

55AX score / 100

Needs work

Readiness diagnosis

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

重大 0件

6 warnings

合格 20件

Current constraintAgent Integration6%

Five readiness layers

Readiness orbit

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

Evidence profile

26 checks recorded

Passed vs. material gaps

Critical

Warnings

Passed

Discovery

4.6 / 13 weighted points

36%

Identity

14.2 / 22 weighted points

65%

Access

28.3 / 31 weighted points

91%

Agent Integration

Current constraint

1.4 / 24 weighted points

User Experience

6.7 / 10 weighted points

67%

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

エージェントインタラクション

Review

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

Actionability

Review

No form or direct contact path — agents have no way to convert here

Actionability

Review

Missing semantic HTML: <header>

コンテンツ品質

Review

Only 33% of images have alt text (6/18)

コンテンツ品質

Review

Explore the audit

Findings and fixes

エージェントインタラクション

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 4 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

No form or direct contact path — agents have no way to convert here
Add an agent-usable conversion path — a contact form, a booking widget, or a Dashform AI funnel agents can complete on the visitor's behalf.

Missing semantic HTML: <header>
Add missing semantic HTML elements: <header>

Only 33% of images have alt text (6/18)
Add descriptive alt attributes to all <img> tags. Example: <img alt="Description of image content" src="..." />

Sufficient text content available for AI extraction

15 social media links 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.

Limited contact information — add phone, email, and address

Response time is 3.2s — slower than agents prefer

Lean HTML payload (459KB)

Response is compressed (br)

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 (HSTS 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

loft.sh

https://loft.sh

Readiness orbit

Top Priorities

Findings and fixes

All Findings

Turn audit findings into an agent-ready funnel.

Related AX Scores