Open source · MIT

Open Agent Readiness

open-agent-readiness scores how well a website works for AI agents: assistants that browse on a person's behalf, coding agents, and MCP clients. Checks run over plain HTTP, with no headless browser. It returns a score from 0 to 100, a letter grade, per-category scores, and a ranked list of fixes, each with a pasteable coding-agent prompt. Protocols that agents do not use yet, such as x402 and A2A agent cards, are detected and shown but never scored.

The scanner powers Agent Readiness on StartupHub.ai. Scoring follows rubric web/2026.11 and is the same scanner the website runs, with method published separately in the agent-readiness-spec repository. Code is MIT licensed.

npx open-agent-readiness scan yoursite.com

Scan in your browser Source code Scoring method npm

What a scan returns

A real scan of anthropic.com, 28 October 2026.

Terminal

  https://www.anthropic.com/
  Agent readiness 57/100 D  rubric web/2026.11

  Discoverability  ████████░░   76   2/5 checks passed
  Content          █████░░░░░   53   3/6 checks passed
  Access Control   █████████░   85   2/3 checks passed
  Capabilities     ██░░░░░░░░   18   0/3 checks passed
  Commerce         ████░░░░░░   40   1/3 checks passed
  Quality          ███████░░░   67   5/7 checks passed

  Fix first (most impact at the top)
  ✗ MCP Server Card  No MCP server card. Agents (Claude Desktop, Cursor) can't discover your tools.
  ✗ Markdown content negotiation  Homepage only returns HTML on Accept: text/markdown. Agents waste tokens on markup.
    ...

What it checks

Category	Checks
Emerging (shown, not scored)	A2A agent card, Agent Skills, agentic registration, API catalog (RFC 9727), x402, x402-mesh, MPP and other agent payment protocols, Web Bot Auth, Content Signals, RateLimit headers
Discoverability	robots.txt, sitemap, Link header (RFC 8288), llms.txt discovery link, AGENTS.md or CLAUDE.md, on-site search URL
Content	Markdown content negotiation, llms.txt and its link notes, llms-full.txt, HTML size within agent limits, page text present in the HTML, facts near the top
Access control	Browse bots can fetch the site, AI bot rules in robots.txt, what each AI and search client actually receives
Capabilities	MCP Server Card and live MCP endpoint, OpenAPI spec, OAuth discovery (RFC 8414, RFC 9728)
Commerce	Pricing and checkout pages reachable, Offer and Product structured data, payment links
Quality	OpenGraph, single H1, image alt text, meta description, canonical URL, JSON-LD and entity JSON-LD, security.txt (RFC 9116), JSON content negotiation

In CI

Fails the build when the score drops below your threshold.

.gitlab-ci.yml

agent-readiness:
  image: node:20
  script:
    - npx open-agent-readiness scan https://example.com --min-score 70

As a library

Node.js 18 or later.

TypeScript

import { scan } from 'open-agent-readiness';

const result = await scan('https://example.com');
console.log(result.total_score, result.grade);

Hosted version

StartupHub.ai has scanned more than 90,000 sites with this scanner. The hosted version at startuphub.ai/agent-readiness adds benchmarks against other sites, score history with alerts when a score drops, the Live Agent Run where real agents attempt tasks on the site, and score badges. Scanning a site there is free.

Scan a site free

MIT licensed. The scoring method is published separately under CC BY-ND 4.0.