Website reader API and MCP for AI agents
TollNode gives Claude, Cursor or any AI agent a whole small site or docs section as clean Markdown (up to 10 pages with title and URL) over API and MCP, from $0.042 per crawl, no subscription.
Ask in plain words, like 'read the x402 docs and summarize how payments work', and get the pages back as text.
No subscription · Sign-up optional · From $0.042 per call · Charged only when it works · No account needed
Read x402.org and summarize what the protocol does
TollNode · website-reader$0.042 · up to 10 pages · 50 s
| Field | Value |
|---|---|
| Page | https://www.x402.org/ |
| Title | x402 |
| Format | Markdown, ready for your agent |
Real result.
What your agent gets from a crawl
Page text
Website to Markdown crawler
Each page as clean Markdown.
- URL
- https://www.x402.org/
- Title
- x402
- Markdown
- # x402 x402 is an open, neutral standard...
You get
- markdown: the page text with headings and links
- url: the page address
- title: the page title
- description: the meta description
- canonical URL: the page's own preferred address
Crawl scope
Where it starts and what it follows.
- Start
- https://docs.example.com/
- Pages
- up to 10
- Follows
- links on the same site
You get
- url: where to start
- pages: up to 10 on the same site
- links followed: same-site links only
- order: start page first
- limits: no logins, no JavaScript-only apps
Every result is structured JSON your agent can sort, filter, summarize or turn into a spreadsheet.
What people ask their AI to do with it
AI engineers
Load docs into an agent's context
“Read the Stripe API quickstart section and answer my question about idempotency keys”
1 call
Cost today$0.042
Researchers
Read a whole small site
“Crawl this nonprofit's site and summarize what programs it runs”
1 call
Cost today$0.042
Sales teams
Understand a prospect
“Read this company's site and tell me who they sell to”
1 call
Cost today$0.042
Competitive analysts
Compare pricing pages
“Read these three competitors' sites and compare their pricing tiers”
3 calls
Cost today$0.126
Content teams
Audit a site
“Crawl our help center and list pages that lack a clear title”
1 call
Cost today$0.042
Developers
Feed a RAG index
“Fetch this docs section as Markdown so I can index it”
1 call
Cost today$0.042
Use it as a Website reader MCP server in Claude, Cursor or Codex
Connect once. Then just ask.
Paste this into Claude Code, Cursor or Codex. It connects TollNode and gives your agent a small wallet. Add a few dollars of USDC and you're done.
npx agentcash add https://tollnode.comCopy for your AI agent
Paste this into your agent so it knows what this node does and how to call it.
You can use TollNode's Website reader node. Call POST https://tollnode.com/api/v1/nodes/website-reader with url (required). The price is one flat price per call, charged only when it returns a result. Over MCP, call the "website-reader" tool; use list_models to see prices. Return results as tables when I ask for lists.Price
$0.042
per search (up to 10 pages)
Charged only when a call returns a result. Failed or empty calls cost $0.
Website reader without running your own tools
No subscription, no credit packs
Pay per crawl: $0.042 for every call, with nothing to choose.
No crawler to run, no proxies, no browser
Your agent calls TollNode. You never set up billing or request access with anyone else.
Only charged when it works
If a call returns nothing or fails, it costs nothing.
One connection, much more
The same setup also gives your agent Google search, YouTube, Maps, images and video, so one prompt can chain them.
FAQ
- Is there an official API for this?
- No. This node is TollNode's own crawler: it follows links on one site and returns each page as Markdown.
- Can I use it as a Website reader MCP server in Claude?
- Yes. Add https://tollnode.com/mcp in Claude Code, Claude Desktop or Cursor, and the "website-reader" tool appears with its fields and price.
- How many pages does it read?
- Up to 10 pages on the same site in one call, each with its URL, title, description and the text as Markdown.
- Does it work on every site?
- No. It cannot read pages behind a login, very large sites, or apps that only render with JavaScript.
- How long does it take?
- About 50 seconds for a full 10-page crawl. For one page, url-to-markdown is faster and cheaper.
- Do I need an account with another service?
- No. Your agent calls TollNode, and you never log in to another service or share cookies.
- What if nothing comes back?
- You're not charged. TollNode only collects payment when a usable result is returned, and your agent can try a different input.
Works well with
Ready in a minute
Connect your AIFor developers: endpoint, inputs and an example response
POST https://tollnode.com/api/v1/nodes/website-reader
Input
| Field | Type | Notes |
|---|---|---|
| url* | string | Where to start, e.g. https://docs.example.com/; links on the same site are followed |
Example response
{
"results": [
{
"url": "https://www.x402.org/",
"markdown": "# x402\n\nx402 is an open, neutral standard for internet-native payments. It absolves the Internet’s original sin by na…",
"metadata": {
"title": "x402",
"author": null,
"keywords": null,
"openGraph": [
{}
],
"description": "x402 is the internet's payment standard. An open standard for internet-native payments that empowers agentic payments…",
"canonicalUrl": "https://x402.org/",
"languageCode": "en-US"
}
}
]
}Also described in the OpenAPI file and llms.txt. Setup for every app: Agents & MCP.