AX監査レポート

https://somosinternet.co

August 12, 20262345ms response

71AX score / 100

良好

Readiness diagnosis

This site is mostly ready for AI agents. A few focused improvements can close the remaining gaps.

重大 0件

3 warnings

合格 29件

Current constraintAgent Integration44%

Five readiness layers

Readiness orbit

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

Evidence profile

32 checks recorded

Passed vs. material gaps

Critical

Warnings

Passed

Discovery

9.3 / 13 weighted points

72%

Identity

17.2 / 22 weighted points

78%

Access

25 / 31 weighted points

81%

Agent Integration

Current constraint

10.5 / 24 weighted points

44%

User Experience

8.7 / 10 weighted points

87%

Start here

Top Priorities

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

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

Actionability

Review

Very large HTML payload (2610KB) — agents may truncate it

Performance

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 4 JSON-LD structured data blocks

Business schema found: Product

Schema has 2 fields — consider adding more for better AI comprehension

OpenGraph tags complete (title, description, image)

Meta tags complete (title, description, canonical)

llms.txt found with correct format

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

RSS/Atom feed detected

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

Priced offers found, but no clear checkout path for agents

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

CTA buttons have descriptive labels

HTML semantic structure is well-formed

82% of images have alt text (14/17)

Sufficient text content available for AI extraction

FAQ schema detected

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

6 social media links found

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

Location or rating data found in schema — consider linking to Google Business Profile

Contact information detected (phone, email, address)

Response time is 2.3s — acceptable, could be faster

Very large HTML payload (2610KB) — agents may truncate it
Trim your initial HTML payload — defer non-critical markup and avoid inlining large data blobs. Agents truncate very large documents.

Response is compressed (br)

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

Entry URL redirects before serving — adds latency for agents

Author/publisher identity declared in schema

Review/rating schema found — agents can cite social proof

Partial security headers (HSTS only)

Content dates (dateModified/datePublished) present in schema

Sitemap shows content updated within the last 6 months

Footer/copyright year is current

HTTPS enabled

SSL certificate valid

No privacy policy link detected

No terms of service link detected

somosinternet.co