---
title: Agent Ready MCP server
description: Run agent-readability scans from Claude Desktop, Claude Code, Cursor, Cline, VS Code, or Windsurf. Hosted Streamable HTTP endpoint, npm package, registry listings on Smithery and the official MCP registry.
last_updated: 2026-09-02
canonical_url: https://agent-ready.dev/mcp
---

# Agent Ready, in your MCP client

> Run agent-readability scans from Claude Desktop, Claude Code, Cursor, Cline, VS Code, or Windsurf. Same scanner, three install paths — the npm stdio server works keyless.

## Install paths

All three call the same scanner. The npm stdio server (`agent-ready-mcp`) works without a key — `scan_site` runs on the anonymous free tier (3 scans/30 days per IP, 25-page depth). A Pro API key from the [dashboard](https://agent-ready.dev/dashboard/api-keys) unlocks deeper scans, history (`get_scan`), and the authenticated hosted endpoint.

### Remote (HTTP)

Lowest-latency option for clients that speak Streamable HTTP MCP natively (Claude Desktop, Cursor, the Vercel MCP consumer).

```json
{
  "mcpServers": {
    "agent-ready": {
      "url": "https://agent-ready.dev/api/v1/mcp",
      "headers": {
        "Authorization": "Bearer ar_live_..."
      }
    }
  }
}
```

### Local (npm)

The published [`agent-ready-mcp`](https://www.npmjs.com/package/agent-ready-mcp) package is a thin stdio wrapper around the hosted REST API. Same key delivered via `AGENT_READY_API_KEY` env var.

```json
{
  "mcpServers": {
    "agent-ready": {
      "command": "npx",
      "args": ["-y", "agent-ready-mcp@latest"],
      "env": {
        "AGENT_READY_API_KEY": "ar_live_..."
      }
    }
  }
}
```

### mcp-remote bridge

Community shim that exposes a remote HTTP MCP server as a local stdio process. Use this if your client only speaks stdio and you don't want to install a per-server package.

```json
{
  "mcpServers": {
    "agent-ready": {
      "command": "npx",
      "args": [
        "-y", "mcp-remote",
        "https://agent-ready.dev/api/v1/mcp",
        "--header", "Authorization:Bearer ar_live_..."
      ]
    }
  }
}
```

## Where is Agent Ready listed?

The MCP server is published across the major MCP registries and marketplaces.

| Channel | Listing |
|---|---|
| Smithery | [agent-ready/agent-ready-mcp](https://smithery.ai/servers/agent-ready/agent-ready-mcp) |
| MCP registry | [io.github.mlava/agent-ready-mcp](https://registry.modelcontextprotocol.io/v0/servers?search=agent-ready-mcp) |
| npm | [agent-ready-mcp](https://www.npmjs.com/package/agent-ready-mcp) |
| GitHub | [mlava/agent-ready-mcp](https://github.com/mlava/agent-ready-mcp) |
| mcp.so | [agent-ready/mlava](https://mcp.so/server/agent-ready/mlava) |
| Glama | [mlava/agent-ready-mcp](https://glama.ai/mcp/servers/mlava/agent-ready-mcp) |
| mcpservers.org | [mlava/agent-ready-mcp](https://mcpservers.org/servers/mlava/agent-ready-mcp) |

## What does the MCP server expose?

Three tools, three prompts, one Claude Skill descriptor. All advertised in the discovery manifest at [`/.well-known/mcp.json`](https://agent-ready.dev/.well-known/mcp.json).

| Kind | Name | Purpose |
|---|---|---|
| tool | `scan_site` | Run a fresh scan against a URL. |
| tool | `get_scan` | Fetch a previous scan by id. |
| prompt | `scan` | Fresh scan + high-level summary. |
| prompt | `interpret_scan` | Plain-English explanation of a scan's findings. |
| prompt | `remediation_plan` | Prioritised fix-it doc (optional `focus`: "seo" or "agents"). |
| resource | `methodology` | Scoring formula, rating bands, JS-rendering handling. |
| resource | `checks` | Reference table of all 72 check IDs. |
| resource | `llms.txt` | Agent Ready's own /llms.txt mirrored. |
| resource | `specs` | Canonical URLs for the specs we validate. |
| skill | `agent-ready` | Anthropic Claude Skill descriptor for activation routing. |

## How do the three install paths differ?

All three call the same scanner with the same API key. The choice comes down to which transport your MCP client supports best.

- **Remote (HTTP)** — point the client at `https://agent-ready.dev/api/v1/mcp` with a Bearer header. Lowest-latency option for clients that speak Streamable HTTP MCP natively.
- **Local (npm)** — spawn `npx -y agent-ready-mcp@latest` as a stdio child process. Same key delivered via `AGENT_READY_API_KEY` env var.
- **mcp-remote bridge** — community shim that exposes a remote HTTP MCP server as a local stdio process. Use this if your client only speaks stdio and you don't want to install a per-server package.

## Frequently asked questions

### What's the difference between the hosted endpoint and the npm package?

Both call the same scanner. The hosted endpoint at agent-ready.dev/api/v1/mcp speaks Streamable HTTP and requires a Pro API key — Claude Desktop and Cursor connect to it directly. The npm package (agent-ready-mcp) is a stdio wrapper that any MCP client can spawn locally; it works without a key (scan_site runs on the anonymous free tier — 3 scans per 30 days per IP at 25-page depth) and uses a Pro key when one is set for deeper scans and history. Pick HTTP for native clients, npm for stdio-only clients (some IDE plugins, Cline) or a zero-config keyless start.

### Which MCP clients can I use this with?

Claude Desktop, Claude Code, Cursor, VS Code (with the MCP extension), Cline, Windsurf, and any other client that speaks the Model Context Protocol. The hosted HTTP endpoint covers clients that support remote MCP servers; the npm package covers stdio-only clients via npx.

### What tools, prompts, and resources does the server expose?

Three tools — scan_site (run a fresh agent-readability scan against a URL), get_scan (fetch a previous scan by id), and ask (natural-language search over Agent Ready's methodology, check registry, and validated specs). Three prompts — scan, interpret_scan, remediation_plan — that wire the tools into end-to-end workflows. Four resources — methodology (the scoring formula), checks (table of all 72 check IDs), llms.txt (Agent Ready's own /llms.txt), and specs (canonical URLs for the specs we validate). One Claude Skill descriptor (skills/agent-ready/SKILL.md inside the npm package) for activation routing.

### Do I need a paid plan?

Yes. The MCP server uses the same Bearer-token authentication as the REST API and only accepts Pro keys. Free-tier scans are available through the web UI but not via the MCP server. Issue a key from your dashboard after upgrading.

### Where can I find the source and report issues?

GitHub at github.com/mlava/agent-ready-mcp. The repo includes the full source, test suite, build pipeline, manifest.json (Glama Marketplace), server.json (MCP registry), and glama.json (Glama maintainer claim). MIT-licensed.

## Related

- [REST API & CI/CD](https://agent-ready.dev/docs/api) — same scanner via REST endpoints or the GitHub Action.
- [Scoring methodology](https://agent-ready.dev/methodology) — how the 72 checks combine into the 0–100 score.

## Sitemap

See the full [sitemap](https://agent-ready.dev/sitemap.md) for all pages on agent-ready.dev.