agent-ready-scanner

Command-line client for Agent Ready — scan any URL for AI-agent readability against the Vercel Agent Readability Spec, the llmstxt.org standard, and agent-protocol manifests (MCP server cards, A2A, agents.json, agent-permissions.json, UCP, x402, NLWeb), plus a separate accessibility sub-score (WCAG 2.2 / layout stability).

It's a thin wrapper over the hosted agent-ready.dev REST API — no scanning happens locally. For tool-native access from an AI assistant, see agent-ready-mcp instead.

Install

npm install -g agent-ready-scanner

This installs the agent-ready command. Or run without installing:

npx agent-ready-scanner scan https://example.com

Why agent-ready-scanner? The bare agent-ready name is blocked by npm's package-name similarity policy (it collides with an unrelated agentready package). The installed command is still agent-ready.

Requires Node.js ≥ 20.10.

Authentication

scan works without a key on the free anonymous tier — 3 scans per 30 days per IP, at 25-page depth:

npx agent-ready-scanner scan https://example.com   # no key, no account

A Pro API key unlocks 50 scans/month, 250-page depth, scan history (get, list), and weekly monitoring. Issue one from the dashboard, then either:

export AGENT_READY_API_KEY="ar_live_..."
# or pass per-command:
agent-ready scan https://example.com --api-key ar_live_...

ask, mcp-scan, and validate-schema are public and need no key.

Commands

scan <url>

Starts a scan, waits for it to finish, and prints a readability summary. Keyless runs use the anonymous free tier (synchronous, 25 pages); with a Pro key the scan runs on the deeper authenticated pipeline with polling.

agent-ready scan https://example.com
agent-ready scan https://example.com --page-limit 25
agent-ready scan https://example.com --no-wait      # queue only, print the id
agent-ready scan https://example.com --json         # raw JSON

get <id>

Fetch a scan by id (e.g. one started earlier with --no-wait).

agent-ready get V1StGXR8_Z
agent-ready get V1StGXR8_Z --json

list

List your recent scans, newest first.

agent-ready list
agent-ready list --limit 5
agent-ready list --cursor 2026-05-30T00:00:00.000Z   # next page

ask <query...>

Natural-language search over Agent Ready's own docs (methodology, the check registry, supported specs). Public — no API key.

agent-ready ask "how is the score calculated?"
agent-ready ask "what does check S4 do?" --type checks
agent-ready ask "summarize the llms.txt requirements" --mode summarize

mcp-scan <endpoint>

Connect to a live, remotely-hosted MCP server and grade its tools, resources, and prompts against MCP best practices — returns a 0–100 MCP quality score with a fix for each issue. Public — no API key. Great for CI (e.g. fail a build when your server scores below a threshold). Remote http(s) endpoints only.

agent-ready mcp-scan https://mcp.example.com/mcp
agent-ready mcp-scan https://mcp.example.com/mcp --json

validate-schema <url|->

Validate a page's JSON-LD structured data — schema lint plus the agent-coherence checks the first-party validators (validator.schema.org, Rich Results Test) don't do (freshness honesty, canonical/.md coherence, entity-name consistency, extraction signal). Pass a URL to fetch and validate, or - to read a JSON-LD string from stdin (paste mode — handy for validating JSON-LD an agent just wrote, no deploy needed). Public — no API key. Exits non-zero when any check fails, so it drops into CI.

agent-ready validate-schema https://example.com/product
cat page.jsonld | agent-ready validate-schema -
echo '{"@context":"https://schema.org","@type":"Product","name":"X"}' \
  | agent-ready validate-schema - --json

Global options

Environment variables

Exit codes

--json output goes to stdout; progress and errors go to stderr, so you can safely pipe JSON into other tools.

Development

npm install
npm test          # vitest
npm run typecheck # tsc --noEmit
npm run build     # bundle to dist/cli.mjs

License

MIT © Agent Ready