📖URL Reader (Markdown)
Fetch a web page server-side and convert it to clean markdown — headings, links, lists, tables and text, with scripts, styles and boilerplate stripped. The 'reader' pattern agents use to actually understand web pages. SSRF-guarded with redirects re-validated, 10s timeout, 1MB cap.
🔌 Server-side tool
This tool runs on the server (it fetches third-party URLs directly and cannot run in your browser). Use the paid JSON API below — the worked examples show real server-side results and refresh automatically.
Worked examples
Read example.com
GET /api/v1/dev/url-reader?url=https%3A%2F%2Fexample.com&maxChars=5000
Result: This domain is for use in documentation examples without needing permission. This is not a service; avoid relying on it for testing and monitoring purposes.
Read a docs page
GET /api/v1/dev/url-reader?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FProof_of_work&maxChars=8000
Result: [Jump to content](#bodyContent) From Wikipedia, the free encyclopedia System that regulates the formation of blocks on a blockchain **Proof of work** (also written as **proof-of-work**, and abbrev…
Machine API (x402)
$0.010 / callThis tool is also a JSON API for AI agents. Requests without payment receive 402 Payment Required plus instructions; agents pay USDC on Base via the x402 protocol — no accounts, no API keys.
GET /api/v1/dev/url-reader?url=https%3A%2F%2Fexample.com&maxChars=5000 HTTP/1.1
Host: localhost:3000
→ 402 (payment required, instructions in headers)
→ 200 (after X-PAYMENT header; JSON body below)
{
"tool": "dev/url-reader",
"input": {"url":"https://example.com","maxChars":5000},
"result": { "value": 156, "answer": "This domain is for use in documentation examples without needing permission. This is not a service; avoid relying on it for testing and monitoring purposes." }
}Agent docs: /llms.txt · OpenAPI spec · integration guide
About this tool
The reader pattern: agents fetch this endpoint with a URL and receive the page's actual content as clean markdown — headings preserved, boilerplate gone — instead of raw HTML soup. Server-side fetching with an SSRF guard (private networks blocked, redirects re-validated) and hard size and time caps.
Frequently asked questions
How is this different from url-metadata?
url-metadata returns the page's *summary* (title, description, OG image). url-reader returns the page's *full content* as markdown for actual reading.
Does it render JavaScript?
No — it converts the served HTML. Static pages and server-rendered sites work fully; client-rendered apps return little text and the tool says so rather than pretending.