AI assistants and agents read websites differently from people. Most of them do not run JavaScript, they look for files such as llms.txt, and they follow robots.txt rules written for them. Agents that do work for someone can also call your API or pay for it, if they can find it. This free scan checks the ten things agents look for and tells you what to fix.
Free, no account. 10 scans per IP address per day. We fetch your public page and a few public files once, identify ourselves as ManHourPlan-AgentReadyScan/1.0, and do not store the results.
| Check | Why it matters | Points |
|---|---|---|
| Readable without JavaScript | Most AI crawlers read the first HTML response only. A JavaScript-only page looks empty to them. | 15 |
/llms.txt | A short Markdown map of your site for language models (llmstxt.org). | 15 |
/llms-full.txt | Optional full text of your key docs in one file. | 5 |
| robots.txt rules for AI crawlers | Whether GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-User, PerplexityBot, Google-Extended and others may read the page. | 10 |
| XML sitemap | Lets crawlers find every page. | 10 |
| Structured data (schema.org) | JSON-LD tells machines what the page is: an organisation, product, service, article or FAQ. | 10 |
/.well-known/mcp.json | Points agents to your MCP server, if you have one. | 10 |
/.well-known/x402.json | Lists API calls agents can pay for per call in USDC, if you sell any. | 10 |
/.well-known/api-catalog | RFC 9727 list of your APIs and their docs. | 5 |
| OpenAPI description | Lets an agent call your API without guessing. | 5 |
The MCP, x402, API-catalog and OpenAPI checks only matter if you have an API or data to offer. A brochure site can be in good shape without them. The scan checks public files and the raw HTML; it cannot tell you whether any AI assistant will cite you, and nobody can promise that.
We set up manhourplan.com this way first. On 22 November 2026 it scored 100 out of 100 on this scan. You can check it yourself: scan manhourplan.com. It has:
What we do
llms.txt (and llms-full.txt where it helps) for your site./.well-known/api-catalog and link your OpenAPI file. If you already run an MCP server or paid API: add mcp.json and x402.json discovery files.What you provide: your website URL, the key pages you care about, and either edit access (CMS, Git repository or hosting) or a developer who can upload the files we send.
Timeline: we aim to finish within 5 business days after we have access.
Not included: writing new page content, building an MCP server or API (see below), and any promise of AI citations or rankings.
An MCP server lets AI assistants such as Claude, ChatGPT and Cursor use your product directly. We build one remote MCP server for one API or data source you own.
/.well-known/mcp.json file and an official MCP Registry entry under your domain.What you provide: API documentation, a test API key or test account, and access to add files to your domain (for the registry proof).
Timeline: we aim to deliver within 10 business days after we have the docs and test access. We confirm the scope by email before starting.
Not included: building the underlying API, OAuth/SSO, and ongoing hosting costs.
x402 lets AI agents pay for an API call with USDC on the spot, without signing up. It is how our own PageMD and Wallet Doctor APIs charge per call.
/.well-known/x402.json catalog.What you provide: the endpoint URLs and prices, a public wallet address to receive USDC (never a private key or seed phrase), and your own Coinbase Developer Platform API key, which you add yourself.
Timeline: files and guide delivered within 3 business days after we have your endpoints.
Not included: changing your API's code, deploying on your servers for you, and any promise of agent traffic or a Bazaar listing.
Payments are processed by Stripe. After you pay, we email you at the address you used at checkout to collect what we need. Questions before buying: support@manhourplan.com.