Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
com.agentgrade avatar

AgentGrade

AgentGrade is an MCP server that scans websites for AI agent readiness, payment protocols, and discovery endpoints.

About

AgentGrade is a remote MCP server that scans websites and reports how ready they are for AI agents, including payment protocols and discovery endpoints. developers shipping APIs, marketplaces, or agent-facing SaaS can point the tool at their own staging URL or a competitor to see gaps before launch. It complements human SEO with agent-oriented visibility checks aligned to how Claude, ChatGPT search, and other systems might route users to actionable surfaces. You add the AgentGrade remote in MCP settings and invoke scan tools from your agent rather than running a local audit repo. Results inform fixes to metadata, well-known agent routes, and payment exposure. It diagnoses readiness; it does not implement fixes or host your app.

  • Scans arbitrary websites for AI agent readiness signals
  • Detects payment protocols relevant to agent checkout flows
  • Surfaces discovery endpoints agents and catalogs may consume
  • Hosted streamable-http MCP at agentgrade.com/mcp
  • Useful competitive benchmark during launch prep

AgentGrade by the numbers

  • Exposes 4 verified tools (MCP introspection)
  • Data as of Jul 7, 2026 (Skillselion catalog sync)
terminal
claude mcp add --transport http mcp-server https://agentgrade.com/mcp

Add your badge

Show developers this MCP server is listed on Skillselion. Paste this into your README.

Listed on Skillselion
TransportHTTP
AuthNone
Tools4

What it does

Audit any site for AI agent readiness, payment protocols, and machine-readable discovery endpoints before you rely on agent traffic.

Who is it for?

Best when you're launching agent-native or API-first products and want a quick external readiness scan from inside your coding agent.

Skip if: Deep penetration testing, compliance attestations, or teams that only need traditional Lighthouse SEO scores.

What you get

After scanning, you get a structured readiness picture you can turn into concrete launch tasks for discovery URLs and agent payment hooks.

  • Readiness findings on agent discovery and protocols
  • Actionable gaps versus agent-capable competitors
  • Launch checklist inputs for GEO and agent routing

By the numbers

  • MCP server version 1.0.0 in published metadata
  • Remote endpoint https://agentgrade.com/mcp via streamable-http
  • Scan scope described as AI readiness, payments, and discovery endpoints

Tools 4

Public tool metadata - what this server can do for an agent.

scan_url1 param

Scan a URL for agent capabilities and payment protocols. Returns full results including x402, MPP, L402, discovery, bazaar, MCP, plugins, OpenAPI, and more.

  • urlstringrequiredURL to scan (http or https)
scan_compact1 param

Compact scan returning only rails, capabilities, and a numeric score. Optimized for agent decision-making.

  • urlstringrequiredURL to scan
get_history2 params

Get scan history for a URL. Requires database to be configured.

  • urlstringFilter by URL (optional)
  • limitnumberMax results (default 20, max 100)
validate_x402_json1 param

Validate x402.json content against expected schema. Returns errors, warnings, and suggestions. No network requests — pure validation.

  • contentobjectrequiredThe x402.json content to validate

Recommended MCP Servers

How it compares

Agent-readiness scanner MCP, not a code-review skill or hosting control plane.

FAQ

Who is AgentGrade for?

Developers and small teams shipping web products that expect traffic or transactions from AI agents and assistant search.

When should I use AgentGrade?

Use it before launch or a major GEO push when you want to verify discovery endpoints and payment exposure on your site or a competitor.

How do I add AgentGrade to my agent?

Configure the remote MCP URL https://agentgrade.com/mcp as streamable-http in Claude Code, Cursor, or another MCP client, then run scan tools against target URLs.

Developer Toolsagentsautomation

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.