Agent-Ready Web Benchmark

See what AI agents can read and do on your site.

Arachne checks machine-readable signals, then tests whether a bounded agent can actually understand your offer, find the right path, and stop before unsafe actions.

Free score No signup Combined scan $99 Implementation Pack Hosted Shadow API path

Run the free Agent Readiness Scan

Enter a public website URL once. Arachne checks machine-readable signals, then automatically tests a bounded agent task and builds one combined result.

No signup. No card. Just the score.

By submitting a URL, you agree that Arachne may scan publicly accessible pages and signals to generate an Agent Readiness result. Arachne does not log in, bypass access controls, or perform write actions without authorization. See our Privacy Policy.

--

Agent Readiness Score

Static scan findings

Agent task result

Top 3 issue categories

    Free recommendation
    Paid report · Full access

    Agent Readiness Implementation Pack

    --
    Today
    with Arachne
    --
    After Shadow API

    Every signal, scored

    Your prioritized fix list

    The fix, done for you

    Don't do the homework. Let Arachne compile it.

    Arachne compiles your site into a Shadow API — a native MCP server, an OpenAPI surface, and an agent policy — without you touching your codebase. Preview the exact Shadow API we'd build for you, right now.

    Make it live

    $500one-time setup + $99/mo hosted
    • Setup: compile, domain-ownership verification, initial tuning
    • Monthly: hosting, re-sync as your site changes, hash-chained metering ledger

    We'll verify your domain and get your Shadow API live. No charge until your build is scheduled.

    Quick answers

    The questions buyers ask most often before clicking unlock.

    Doesn't this make my site more vulnerable to AI agents?

    Agents and scrapers can already read public pages. Arachne turns that invisible traffic into a governed, logged interface.

    Publishing a Shadow API does not grant blanket write access. Reads are structured, writes default to draft_only, high-risk actions require approval, and page content is treated as data rather than instructions. Every call lands in a tamper-evident ledger.

    The trade is unmanaged scraping versus accountable access.

    How is my score calculated?

    Deterministic, 7 weighted signals, max 100. Each is binary — full weight or zero. The weights sum to 100.

    SignalWeight
    MCP server endpoint35
    OpenAPI spec at root25
    llms.txt15
    sitemap.xml10
    robots.txt5
    JSON-LD on homepage5
    Semantic forms5

    Grade: A (90+) · B (75+) · C (60+) · D (40+) · F (<40)

    The $99 Implementation Pack breaks down exactly which signal you missed, why it matters, and your projected score after Arachne compiles a Shadow API.

    What happens after I pay the $99?

    Instant redirect to your Implementation Pack — every readiness signal scored with the specific detail, a prioritized fix list, and your projected score after Arachne. Plus a live preview of the Shadow API we'd compile for your domain (real manifest, real tools), and the option to commission a hosted build.

    Is the $99 a subscription?

    No. One-time, per domain. You only enter subscription territory if you commission a hosted Shadow API ($500 setup + $99/mo), and that's an explicit separate decision on the next page.

    What MCP client do I need?

    The Shadow API works with any MCP-capable agent. Most common: Claude Desktop, plus Claude Code, Cursor, Continue, and custom agents built with LangChain MCP adapters or the OpenAI Agents SDK. One URL, every client.

    Do you need access to my codebase?

    No. Arachne is fully egress-side — we only see what a normal browser sees on your public site. Domain ownership is proven via a single DNS TXT record. We never touch your code, your database, or your auth system.

    More questions? See the full Arachne FAQ or email [email protected].

    
    ask iris
    IRIS ONLINE
    Drop files here Images, PDFs, text files