AI Agent Readiness — Integrate in 60 seconds
Scan any public URL against 18 agent-readiness standards. Free. No auth. Available via REST, GET one-liner, and MCP.
Try it
REST API
curl -sX POST https://www.startuphub.ai/api/agent-readiness/scan \
-H 'Content-Type: application/json' \
-d '{"url":"https://your-site.com"}'Or GET:
curl 'https://www.startuphub.ai/api/agent-readiness/scan?url=https://your-site.com'Rate limit: 10 scans / 10 min / IP. No auth.
MCP (Claude Desktop, Cursor, Windsurf)
Add this to your MCP config:
{
"mcpServers": {
"startuphub-agent-readiness": {
"command": "npx",
"args": ["mcp-remote", "https://www.startuphub.ai/api/v1/mcp"]
}
}
}Then prompt your agent:
“Call thescan_agent_readinesstool onhttps://example.comand tell me the top 3 fixes.”
Discovery card: /.well-known/mcp/server-card.json
Response shape
{
"total_score": 72,
"grade": "B",
"categories": [
{ "category": "discoverability", "score": 100, ... },
{ "category": "content", "score": 60, ... },
...
],
"checks": [
{ "id": "robots_txt", "status": "pass", "title": "robots.txt present", ... },
...
],
"extras": {
"tokens_html": 53201,
"tokens_markdown": 12740,
"token_savings_pct": 76,
"ttfb_ms": 145,
"wins": ["MCP Server Card — <0.01% of sites have one"],
"weaknesses": ["Markdown content negotiation", "llms.txt guide"]
},
"markdown_summary": "# Agent-Readiness Report..."
}What we check
- Discoverability: robots.txt, sitemap, Link header (RFC 8288)
- Content: markdown content negotiation, llms.txt, llms-full.txt
- Access control: AI bot rules, Content Signals, Web Bot Auth
- Capabilities: MCP Server Card, API Catalog (RFC 9727), OpenAPI spec, Agent Skills, OAuth discovery (RFC 8414 + 9728)
- Commerce: x402 Payment Required
- Quality (extras): OpenGraph, JSON-LD structured data, RateLimit headers (RFC 9331), security.txt, JSON content negotiation
Each failing check returns a fix prompt
Paste straight into Claude Code / Cursor / Windsurf / Copilot. No rewriting.