Without a key: 5 audits per hour and 20 per day per network. With a key there is no per-network limit and sitemap batches go up to 50 URLs.

AISeeWhy API and MCP server

Q: How do I call the API?

POST JSON with a url (and optionally language, bots, markdown) to https://api.aiseewhy.com/v1/audit. Send an x-api-key header for keyed access; without a key the limit is five audits per hour.

Q: How do I connect an AI agent over MCP?

Add a remote MCP server with the URL https://api.aiseewhy.com/mcp (streamable HTTP) and call the audit_url, audit_sitemap or explain_finding tools.

Q: What does a finding look like?

Each finding has an id, a priority (P1, P2 or P3), the crawler it concerns, a title and detail, the evidence seen and ready-to-paste fixes.

Send one POST with a URL to https://api.aiseewhy.com/v1/audit and get a JSON report: findings with priority, evidence and ready-to-paste fixes, in any of 30 languages. AI agents connect to https://api.aiseewhy.com/mcp over MCP and call audit_url. Without a key you get five audits per hour.

How do I call the API?

curl -s https://api.aiseewhy.com/v1/audit \
  -H 'content-type: application/json' \
  -d '{"url": "https://example.com/pricing", "language": "en", "markdown": false}'

Fields: url (required; https:// is added when missing), language (report language code, default en), bots (optional list of crawler names), markdown (add a Markdown report). With a key, send x-api-key: …; sitemap batches use POST /v1/sitemap with {"sitemap_url": …, "max_urls": 50} and need a key. The OpenAPI description is at https://api.aiseewhy.com/v1/docs.

How do I connect an AI agent over MCP?

Add a remote MCP server with the URL https://api.aiseewhy.com/mcp (streamable HTTP). Tools: audit_url, audit_sitemap (up to 10 pages) and explain_finding. The Model Context Protocol documentation explains how clients add servers.

{
  "mcpServers": {
    "aiseewhy": { "type": "http", "url": "https://api.aiseewhy.com/mcp" }
  }
}

What does a finding look like?

{
  "id": "access.blocked", "priority": "P1", "bot": "OAI-SearchBot", "bot_class": "search",
  "title": "OAI-SearchBot is blocked",
  "evidence": { "status": 403, "control_status": 200, "waf": "cloudflare" },
  "fixes": [ { "kind": "cloudflare", "title": "Let OAI-SearchBot through in Cloudflare", "body": "…" } ]
}

Priorities: P1 means AI answers cannot read or cite the page, P2 is important, P3 is nice to have or a policy choice (for example, blocking a training crawler). robots.txt is read as defined in RFC 9309.

What are the limits?

Access	Limit
No key (website, API, MCP)	5 audits per hour and 20 per day per network
API key	No per-network limit; sitemap batches up to 50 URLs
Batch audits by sitemap	Up to 5,000 URLs per run, paid per audited URL

Requests to private or internal addresses are refused. Pages that cannot be audited are reported with a fetch.error or fetch.status finding.