Skip to content
MCP server

AgenticWebScore

Scores a public website's readiness for AI agents with 60 checks and evidence

Install
Pricing
Free
Platforms
Web, API
Runs
Remotely
Version
0.3.0
Listing updated
3 days ago

About AgenticWebScore

AgenticWebScore is a free MCP server for marketing and seo. It runs as a hosted remote server, so there is nothing to install locally.

Every finding comes with a way to check it. A scan uses a plain HTTP client against a public rubric and returns, for each of 60 checks, its status, the evidence read, the source it follows and a shell command that repeats the measurement. The score weights adopted web standards at 70% and agent conventions at 30%, and each record carries a fingerprint that can be verified.

The checks cover robots.txt and AI crawler rules, sitemaps, headings and structured data, HTTP caching and real 404s, markdown negotiation and llms.txt, and OpenAPI and MCP servers. If a site refuses automated clients the result is reported as not measured rather than as a low score, and nothing earns points that was not observed.

Four tools split reading from scanning. scan_site runs a scan, reusing a stored record younger than 6 hours unless fresh is set, while get_score, list_checks and get_report only read stored data. Scans send only GET requests, apart from an MCP handshake when the site runs a server, and never call that server's tools. Limits are 10 new scans per client per minute, 3 at once and up to 60 seconds each.

Install

Run in a terminal

claude mcp add --transport http scan https://agenticwebscore.com/mcp

Add to Settings → Connectors → Add custom connector

https://agenticwebscore.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "scan": {
      "url": "https://agenticwebscore.com/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "scan": {
      "type": "http",
      "url": "https://agenticwebscore.com/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "scan": {
      "serverUrl": "https://agenticwebscore.com/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "scan": {
      "type": "http",
      "url": "https://agenticwebscore.com/mcp"
    }
  }
}

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

scan_site
Scans a public domain and returns the score, level, check evidence and fingerprint.
get_score
Returns the stored record for a host without starting a scan.
list_checks
Lists the 60 checks with kind, module, dimension and rationale.
get_report
Returns the markdown report of a stored record or one check.

What it does

60-check rubric
Each check lists status, evidence, source and a repeatable command.
Weighted score
70% web standards and 30% agent conventions, with a verifiable fingerprint.
Not-measured outcomes
Blocked sites are not scored down.
Stored records
Read earlier scores and reports without starting a scan.
Cautious fetching
Only GET requests, and redirects to private addresses are refused.
No sign-up
The hosted endpoint needs no key.

Good for

  • Audit a site's llms.txt and robots.txt rules for AI crawlers
  • See whether a site's MCP server responds to a handshake
  • Compare a site's score against listed examples
  • Get a markdown report of failed checks to fix

Pricing

Free

Scanning is free; no key required

Prices change. Check agenticwebscore.com before you buy.

AgenticWebScore FAQ

Does it need an API key?

No. The README says there is no key and no sign-up, and scanning is free.

Can a scan hurt my site?

The server sends only GET requests, plus an MCP handshake if you run an MCP server, and it never calls that server's tools.

How current are results?

A stored record younger than 6 hours is returned unless you ask for a fresh scan.

What are the limits?

10 new scans per client per minute, 3 running at once, and up to 60 seconds per scan.

How do I add it to Claude Code?

Run claude mcp add --transport http agenticwebscore https://agenticwebscore.com/mcp.

Is AgenticWebScore free?

Yes. AgenticWebScore is free to use.

How do I add AgenticWebScore to Claude Code?

Run this command in your terminal: claude mcp add --transport http scan https://agenticwebscore.com/mcp

What are the best alternatives to AgenticWebScore?

The closest alternatives to AgenticWebScore are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to AgenticWebScore

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools