Free AI Infrastructure Tool

Is Agentic: AI Agent Readiness
& Diagnostic Site Auditor

Score how easily autonomous AI agents can discover, access, and interact with your site or application. Inspect 5 funnel layers and generate a one-click Prompt-to-Fix for Cursor and Claude Code.

  • 5-layer diagnostic funnel
  • Discovery, APIs, MCP & Protocols
  • Prompt-to-Fix for agents

Making web applications agent-ready

Free diagnostic tool that scores how discoverable, accessible, and actionable a website or API is for autonomous AI agents.

Why it matters

Autonomous AI agents like Claude Code, Cursor, ChatGPT, Hermes, and Antigravity navigate the web differently than human users. They cannot solve complex CAPTCHAs, wait for heavy client-side JavaScript hydration bundles, or decipher ambiguous interactive controls without semantic markup.

Is Agentic evaluates public websites and developer platforms across five essential funnel layers: Discovery, Identity & Auth, Access & APIs, Payments, and Experience. Each audit measures technical baselines including server-rendered HTML without JS, robots.txt bot rules, llms.txt context files, OpenAPI specs, Model Context Protocol (MCP) servers, and agentic micropayment standards (x402, MPP, UCP).

What the tool does

Beyond providing a 0–100 readiness score and letter grade, the auditor extracts concrete technical recommendations and automatically synthesizes a prioritized 'Prompt to Fix'. You can paste this brief directly into your AI coding assistant to resolve all discovered gaps in a single engineering pass.

Terms worth knowing

Agent Readiness Score
A 0–100 benchmark evaluating how reliably autonomous agents can discover, parse, and execute tasks on a website.
llms.txt
A standardized markdown file located at /llms.txt that provides concise, curated architectural context and documentation specifically formatted for LLMs.
Model Context Protocol (MCP)
An open protocol enabling AI models to safely discover tools, resources, and API capabilities exposed by a server.
Agentic Payments (x402 & MPP)
HTTP-native micropayment and checkout protocols allowing autonomous agents to pay for API usage or goods via HTTP 402 Payment Required.
Prompt to Fix
A pre-structured engineering prompt containing all failed and partial checks, formatted for immediate execution in Claude Code or Cursor.

Frequently asked questions

Short answers for the decisions that usually come up after a first run.

What does an Agent Readiness Score measure?

It measures the public accessibility and friction of your website for AI agents: whether content is readable without JavaScript, if robots.txt permits AI crawlers, whether an OpenAPI spec or MCP server is available, and if agentic authentication or payment protocols are supported.

Is the Is Agentic tool free?

Yes. The diagnostic scan, report, and Prompt-to-Fix generation are completely free with no account or API key required.

Does running an audit consume my LLM API tokens?

No. The scoring and heuristic verification are performed directly by the diagnostic engine, consuming zero OpenAI or Anthropic tokens from your accounts.

How can I improve my website's score?

Start by addressing Essential checks: ensure your homepage serves clean semantic HTML before client-side JS loads, add a /llms.txt file with key docs, configure robots.txt to welcome AI user-agents, and expose standard RateLimit headers and an OpenAPI spec.

Who runs the checks?

SquadRoom does. The scan requests your homepage and the standard discovery files (robots.txt, sitemap.xml, llms.txt, the OpenAPI and MCP locations, agent cards) from our servers and scores what comes back. It only scans public websites.