
AgentGrade
from 03prabh
Check is your site agent friendly? Get a badge to prove it.
AgentGrade
Scan any website for AI agent readiness. Get a scored report covering 80+ checks across payment protocols, API discovery, and machine-readable metadata.
Tools
-
scan_url โ Full scan with scored report, capability detection, and consistency checks
-
scan_compact โ Flat JSON optimized for agent decision-making
-
get_history โ Query past scan results
-
validate_x402_json โ Validate x402.json payment discovery files
What It Checks
-
Payment protocols โ x402, MPP, SPT (Stripe), L402 (Lightning)
-
API discovery โ MCP servers, OpenAPI specs, AI plugins, Claude plugins, Skills
-
Machine-readable content โ llms.txt, agents.txt, robots.txt, content negotiation
-
Bazaar discovery โ x402.json service catalogs, discoverable endpoints
-
Homepage & meta โ JSON-LD, OpenGraph, canonical URLs, favicons
-
Infrastructure โ HTTPS, sitemap, cache headers, structured errors
Usage
No auth required. Add to your MCP client config:
{ "mcpServers": { "agentgrade": { "command": "npx", "args": ["-y", "mcp-remote", "https://agentgrade.com/mcp"] } } }
Or use the CLI: npx agentgrade-cli https://example.com
Web scanner: https://agentgrade.com
No common issues documented yet. If you hit a problem, the repository's GitHub Issues page is the best place to look.