mcpField
A
agentfit

agentfit

Local
github.com/CloudCorpRecordsDeveloper Tools0Updated 1w ago

About agentfit

agentfit is an MCP server for requesting AgentReady’s free preview of a website’s readiness for AI agents. It is for site owners or agents checking access, readable content, discovery, and agent interfaces; results are returned from the upstream service without local scoring.

Key features

  1. 01Request a website readiness preview
  2. 02Return a 0–100 score and A–F grade
  3. 03Show category scores for readiness areas
  4. 04Pass through the upstream response

Use cases

  1. 01Check whether a site is accessible to agents
  2. 02Review a site's content readability
  3. 03Inspect discovery and agent-interface scores

MCP support

Its MCP server exposes the `audit_website` tool, which requests the free preview from AgentReady and returns the upstream result. Paid routes and wallet-key payment mode are not implemented.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"cloudcorprecords-agentfit": {
"command": "npx",
"args": [
"-y",
"@reneturcios/agentfit"
]
}
}
}
Run commandnpx -y @reneturcios/agentfit

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@reneturcios/agentfit0.1.0
ocighcr.io/cloudcorprecords/agentfit:0.1.0

Tools

1

Saved tool list · updated 1w ago

ParameterTypeRequiredDescription
urlstringYesURL for this call.

Details

From the project README

agentfit

Is your website ready for AI agents? agentfit asks the AgentReady service for its free preview of a site: a 0-100 score, an A-F grade and category scores covering things like agent access, readable content, discovery and agent interfaces. The score is the upstream's — nothing is computed, trimmed, renamed or summarised locally; the response body comes back exactly as the service returned it.

Free tier only — the paid surface is omitted

AgentReady also offers x402-paid routes (/v1/lite at $0.005 and /v1/audit at $0.03, paid in USDC), and its original stdio server pays for the full audit per call when a wallet private key is supplied through AGENTREADY_X402_PRIVATE_KEY. This rebuild implements none of that. It calls only the documented free route, GET /v1/preview?url=, which is the original server's own default behaviour without the key. The paid routes and the wallet-key mode are omitted entirely: this build holds no payment instrument, never reads AGENTREADY_X402_PRIVATE_KEY or any other wallet or payment environment variable, and never sends an Authorization header.

The free preview is limited upstream to 10 requests per minute per IP; a 429 (like any other non-2xx reply) surfaces as agentready error: HTTP 429.

Configuration

Two environment variables control the client (both are this rebuild's own, labelled):

  • AGENTREADY_API_URL (optional) — API base URL. Defaults to https://api.youraitools.co/agentready; a trailing slash is stripped.
  • AGENTREADY_TIMEOUT_MS (optional) — per-request timeout in milliseconds, enforced with an AbortController. Defaults to 30000.

Tools

  • audit_website — is this website ready for AI agents? Returns the free preview: a 0-100 score, an A-F grade and category scores. Takes one required argument, url (the documented REST parameter; the MCP-layer argument name is a labelled reconstruction, because the hosted MCP endpoint did not answer during spec capture).

An upstream non-2xx reply fails with agentready error: HTTP <status> — that is how the upstream's 400 (bad, private or internal URL), 422 (unresolvable host) and 429 (preview rate limit) answers surface. An unreachable service, including a timeout, fails with agentready unreachable.

CLI

The tool is a subcommand taking one JSON argument string:

npx @reneturcios/agentfit audit_website '{"url":"https://example.com"}'
npx @reneturcios/agentfit --version

Output is the tool's result text (the upstream body on success). Exit codes: 0 on success, 1 on a runtime error, 2 on a usage error.

MCP server

npx -y @reneturcios/agentfit mcp

MCP client config:

{
  "mcpServers": {
    "agentfit": {
      "command": "npx",
      "args": ["-y", "@reneturcios/agentfit", "mcp"]
    }
  }
}

Docker

docker run -i ghcr.io/cloudcorprecords/agentfit
docker run --rm ghcr.io/cloudcorprecords/agentfit audit_website '{"url":"https://example.com"}'

Development

npm install
npm run build   # tsc -> dist/
npm test        # tsx --test test/*.test.ts (fake fetch only, no network)
npm run dev     # run the CLI from source

License

MIT — Copyright (c) 2026 Rene Turcios

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.