CanAgentUse tools

UCP Suite

Validate Universal Commerce Protocol discovery, then test product search, carts, checkout links, and merchant handoff flows.

MCP Playground

Connect to remote MCP servers, inspect tools and resources, test prompts, auth, headers, notifications, and JSON-RPC responses.

A2A Playground

Inspect Agent Cards, validate advertised endpoints, and prepare safe requests for agent-to-agent workflows.

Agent Website Viewer

Enter a public URL and see the roles, names, landmarks, controls, and blockers that shape how AI agents understand the page.

SCANNED
Jul 3, 2026, 3:47 AM
VISIBILITY
Public
Rescan Report
Share Report
Copy Report Link
Export Report
45/100

OVERALL SCORE

Level 2, Agent-Limited

Priority improvements needed for AI agents
AIDiscoverability49%Agent Easeof Use50%Security& Trust66%GEO, AIO, AEO59%SEO100%Performance49%Accessibility0%
  • AI Discoverability 49 out of 100
  • Agent Ease of Use 50 out of 100
  • Security & Trust 66 out of 100
  • GEO, AIO and AEO 59 out of 100
  • SEO 100 out of 100
  • Performance 49 out of 100
  • Accessibility 0 out of 100

CAPTURED SCREENSHOT

Captured website desktop screenshot

What AI sees of your website

NCAA.com – The Official Website of NCAA Championships | NCAA.com favicon

NCAA.com – The Official Website of NCAA Championships | NCAA.com

NCAA.com features live video, live scoring, rankings, news and statistics for all college sports across all divisions in the NCAA.

Open Agent View

Next step

Turn this report into a fix workflow

1 failed check is ready to move into MCP or CLI remediation. Generate a repair prompt, connect the scanner to your coding agent, or open the integration docs before your next rescan.

Fix with MCP / CLI
Detailed report scores grouped by capability area
MetricScoreStatusPassedFailedWarningEvidence
AI Discoverability
49
Priority fix2031View details
Discoverability
47
Priority fix600
Content Readiness
44
Priority fix1100
Bot Access Control
54
Needs work331View details
Agent Ease of Use
50
Needs work110View details
Google Agentic Browsing
50
Needs work110View details
GEO, AIO and AEO
59
Needs work600
GEO Readiness
Not Applicable
Not Applicable200
AIO Readiness
Not Applicable
Not Applicable400
SEO
100
Strong1000
SEO
100
Strong1000
Security & Trust
66
Needs work1600
Security & Trust
66
Needs work1600
Performance
49
Priority fix1200
Performance
49
Priority fix1200
Accessibility
0
Priority fix010View details
Accessibility
0
Priority fix010View details

Prioritized recommendations

Issues ranked by score impact

3 items need attention

AI DiscoverabilityBot Access ControlEmerging recommendation

ai.txt policy

ai.txt policy failed at "Fetch /ai.txt".

82 Fail

Needs attention

ai.txt policy

Failed check
01

Issue

/ai.txt returned HTTP 403.

Details

02

Why it matters

ai.txt is a fragmented emerging convention. It can communicate human-readable AI crawling, training, attribution, restriction, and contact guidance, but it is not a standard access-control mechanism and absence should not be penalized.

Check name

ai.txt policy

Score

18/100

Status

fail

Category

Bot Access Control

Maturity

Emerging recommendation

Goal

Publish an advisory human-readable AI usage policy only when the site intentionally needs one.

Result

ai.txt policy failed at "Fetch /ai.txt".

Validation steps

  1. Fetch /ai.txt

    /ai.txt returned HTTP 403.

    /ai.txt could not be fetched when present
  2. Validate transport

    ai.txt appears to be an HTML page or error document rather than a plain-text policy.

    /ai.txt transport is invalid or not plain text
  3. Parse ai.txt policy

    Transport validation failed before policy parsing could be trusted.

    /ai.txt policy format is missing required sections or non-standard
  4. Validate policy content

    ai.txt does not follow the AI Visibility section model; treating it as non-standard advisory text.

    /ai.txt content is incomplete or unsafe
Evidence log4 steps · 18 lines
Fetch /ai.txt [fail]! /ai.txt returned HTTP 403.INFOFetch /ai.txtINFORequesting optional advisory policy file at /ai.txtFAILCompare /ai.txt HTTP response actual=403 expected="2xx when ai.txt is intentionally published; 404/410 means absent optional file" contentType="text/html" length=150FAIL/ai.txt returned HTTP 403. statusCode=403Validate transport [fail]! ai.txt appears to be an HTML page or error document rather than a plain-text policy.INFOValidate transportINFOChecking ai.txt media type, size, line count, and whether the response looks like plain text contentType="text/html" length=150 lineCount=8FAILCompare media type to preferred text/plain transport actual="text/html" expected="text/plain preferred; readable text required"FAILai.txt appears to be an HTML page or error document rather than a plain-text policy.Parse ai.txt policy [fail]! Transport validation failed before policy parsing could be trusted.INFOParse ai.txt policyINFODetecting ai.txt convention from bracket sections, frontmatter, headings, and ai.txt context linksINFODetected policy structure convention="unknown-text" sections=[] markdownHeadings=[]WARNCheck AI Visibility required sections actual="missing identity, permissions, restrictions" expected="identity, permissions, and restrictions" missingRecommended=["attribution","contact","content-types"]WARNTransport validation failed before policy parsing could be trusted.Validate policy content [warning]! ai.txt does not follow the AI Visibility section model; treating it as non-standard advisory text.INFOValidate policy contentINFOChecking required section content, permission/restriction language, attribution, contact details, and safety risks convention="unknown-text"INFODetected policy signals hasPermissionLanguage=false hasRestrictionLanguage=false hasTrainingLanguage=false hasAttributionLanguage=false hasContact=falseWARNCheck required and recommended section coverage actual={"missingRequired":["identity","permissions","restrictions"],"missingRecommended":["attribution","contact","content-types"]} expected={"missingRequired":[],"recommendedPresentWhenPossible":true}WARNai.txt does not follow the AI Visibility section model; treating it as non-standard advisory text.

Agent Ease of UseGoogle Agentic BrowsingBrowser audit

Accessibility tree is not well-formed

100 Fail

Check name

Accessibility tree is not well-formed

Score

0/100

Status

fail

Device

desktop

Category

Google Agentic Browsing

Fix guidance

A well-formed [accessibility tree](http://goo.gle/lighthouse-agentic-a11y) helps AI agents to navigate and interact with the page.

Evidence
{
  "description": "A well-formed [accessibility tree](http://goo.gle/lighthouse-agentic-a11y) helps AI agents to navigate and interact with the page."
}

AccessibilityAccessibilityBrowser audit

ARIA roles used must conform to valid values

100 Fail

Check name

ARIA roles used must conform to valid values

Score

0/100

Status

fail

Device

desktop

Category

Accessibility

Fix guidance

ARIA roles used must conform to valid values

Evidence
{
  "ruleId": "aria-roles",
  "impact": "critical",
  "description": "Ensure all elements with a role attribute use a valid value",
  "tags": [
    "cat.aria",
    "wcag2a",
    "wcag412",
    "EN-301-549",
    "EN-9.4.1.2",
    "RGAAv4",
    "RGAA-7.1.1"
  ],
  "affectedNodeCount": 2,
  "nodes": [
    {
      "target": [
        ".header-main"
      ],
      "html": {
        "discarded": true,
        "reason": "[discarded fetched document data]",
        "originalType": "string",
        "originalLength": 51
      },
      "failureSummary": "Fix all of the following:\n  Role must be one of the valid ARIA roles: header"
    },
    {
      "target": [
        "footer"
      ],
      "html": {
        "discarded": true,
        "reason": "[discarded fetched document data]",
        "originalType": "string",
        "originalLength": 39
      },
      "failureSummary": "Fix all of the following:\n  Role must be one of the valid ARIA roles: footer"
    }
  ]
}

Sign in to see 44 other issues and the full report

Create a free account to unlock every issue, evidence details, exports, and higher free limits.

Email code

Fix with MCP or CLI

Use this report as the handoff into remediation. Generate a coding-agent prompt with the failing checks attached, or jump to the MCP and CLI setup docs before your next rescan.

Fix with MCP / CLI

Score history

Public scan score over time

Public reports for this website origin. Select any point or report link to open that canonical report.

1 reports
Public score history report links
Scan dateScoreReadinessReport
45/100Level 2, Agent-LimitedCurrent report