Bulk web page reader API
Read thousands of web pages as clean Markdown or text for an LLM: one GET per URL, run as many in parallel as you like, pay only for pages that read.
- Main content as Markdown or plain text with headings, tables, code blocks and links; menus and footers left out
- No rate limit on our side: run a worker pool over your list and keep going
- Bot-blocked, empty, JavaScript-only and missing pages are never charged, so a messy list costs only what it reads
- A whole site from its sitemap: list the URLs, then read them
A free key makes 100 calls a month: leave an e-mail address, open the link, paste the key. Failed calls are never charged.
One request
curl -H "Authorization: Bearer $API_KEY" \
"https://glyphcall.com/extract?url=https%3A%2F%2Fexample.com"
Python
import os, requests
r = requests.get(
"https://glyphcall.com/extract",
params={
"url": "https://example.com",
},
headers={"Authorization": f"Bearer {os.environ['API_KEY']}"},
timeout=60,
)
r.raise_for_status()
print(r.json())
Node.js
const params = new URLSearchParams({
url: "https://example.com",
});
const r = await fetch(`https://glyphcall.com/extract?${params}`, {
headers: { Authorization: `Bearer ${process.env.API_KEY}` },
});
if (!r.ok) throw new Error(await r.text());
console.log(await r.json());
Read a list of URLs
$0.002 a page with a key: $2 for 1,000 pages that read, nothing for pages that fail. Each page is one GET, so a worker pool is all the batching you need.
Measured 18 Nov 2026: 100 current Hacker News links, 8 at a time, read by the same code from a home PC (not from our servers): 82 read, all 100 done in 19 s, median 350 ms a page (90th percentile 774 ms). The 18 that failed (9 sites blocking bots, 6 JavaScript-only pages, 2 timeouts, 1 missing page) would not have been charged.
Python, 8 at a time
import os, requests
from concurrent.futures import ThreadPoolExecutor
KEY = os.environ["API_KEY"]
urls = [line.strip() for line in open("urls.txt") if line.strip()]
def read(url):
r = requests.get("https://glyphcall.com/extract", params={"url": url, "format": "markdown"},
headers={"Authorization": f"Bearer {KEY}"}, timeout=60)
return url, (r.json().get("markdown") if r.ok else None), r.status_code
with ThreadPoolExecutor(max_workers=8) as pool:
for url, md, status in pool.map(read, urls):
print(status, url, len(md or "")) # failed pages are not charged
Node.js, 8 at a time
const urls = (await import("node:fs")).readFileSync("urls.txt", "utf8").split("\n").filter(Boolean);
const read = async (url) => {
const q = new URLSearchParams({ url, format: "markdown" });
const r = await fetch(`https://glyphcall.com/extract?${q}`, { headers: { Authorization: `Bearer ${process.env.API_KEY}` } });
return { url, status: r.status, markdown: r.ok ? (await r.json()).markdown : null };
};
let i = 0;
const results = [];
await Promise.all(Array.from({ length: 8 }, async () => { while (i < urls.length) results.push(await read(urls[i++])); }));
console.log(results.filter((x) => x.markdown).length, "of", urls.length, "read");
A whole site from its sitemap
curl -H "Authorization: Bearer $API_KEY" \
"https://glyphcall.com/sitemap?url=https://www.sitemaps.org/sitemap.xml&maxUrls=100"
# then read each entries[].url as above
The answer
{
"url": "https://agent-data-tools.revmesh2074.workers.dev/examples/extract/sources/short-article.html",
"finalUrl": "https://agent-data-tools.revmesh2074.workers.dev/examples/extract/sources/short-article.html",
"status": 200,
"contentType": "text/html;charset=utf-8",
"sourceFormat": "html",
"linkBaseUrl": "https://agent-data-tools.revmesh2074.workers.dev/examples/extract/sources/short-article.html",
"sourceMetadata": {
"verification": "not_verified",
"authors": [],
"dates": [],
"truncated": false
},
"title": "Owned sample: short article",
"description": null,
"lang": "en",
"canonical": null,
"headings": [
{
"level": 1,
"text": "A short release note",
"id": null
}
],
"section": null,
"links": [],
"tables": [],
"codeBlocks": [],
"contentSelection": "main",
"warnings": [],
"format": "text",
"text": "A short release note\n\nVersion 1.1.1 preserves literal code whitespace and nested list structure.\n\nThe price remains $0.002 per successful extraction. This is an owned sample, not a purchase receipt.",
"textLength": 198,
"truncated": false,
"contentHash": "sha256:6cd3137dc4735c9ab317c764c78b47712dc9ca8eabe435dd0db9142bdaacbe2c",
"hashFormat": "text",
"fetchedAt": "2026-10-16T11:06:01.780Z"
}
Pricing
One API key, paid by card through Stripe. $0.002 per successful call with a key; failed calls cost nothing.
A plan adds its credits each month when the payment succeeds; unused credits stay, and you can cancel at any time. Packs never expire. After paying you get your key on the next page. The same key works on every tool in the catalog. Details: credits and plans.
Inputs
| Name | Type | Description |
|---|---|---|
url (required) | string | Absolute http(s) URL of the page to fetch |
maxChars | integer | Maximum characters of body text to return (500-100000, default 20000) |
format | string | Returned representation: text (default), markdown, or both. HTML structured details are included in every format; Markdown/plain sources have empty details. |
section | string | Optional visible HTML ID within primary content, 1-500 characters. A leading # (the fragment a docs link shows) or a URL ending in #id is accepted, and so is a heading's visible text when exactly one heading matches (answered with warning section_matched_by_heading). Heading scope includes subsections until next same/higher-level heading; section/article/main/div scope is its subtree. Missing/ambiguous/unsupported targets fail uncharged, and a miss lists the ids the page does offer under sections. A fragment added to url still does not select a section. Leave it out to read the whole page. |
What it does not do
- Public pages only: no logins or paywalls.
- Sites that block automated readers stay blocked; the call fails and is not charged.
- The bulk reader does not run JavaScript; single-page apps need the rendered reader, which costs more per page.
Also on the same key
- Sitemap parser, URL extractor: documentation URLs from a bounded XML sitemap · $0.002 a call
- Web page to Markdown, rendered: a JavaScript-rendered page as text or Markdown · $0.01 a call
- Get contents, web contents: clean text of up to five pages in one call · $0.01 a call
- PDF to text, PDF text extraction: extract the text of a PDF document from a URL, page by page · $0.002 a call
Questions
Is there a rate limit?
Not on our side. Each page is one GET; run them in parallel. The sites you read may slow down or block heavy traffic from one reader, so spread a large list over many sites or keep a few requests per site at a time.
What does a failed page cost?
Nothing. Pages that block bots, need JavaScript, are empty, time out or are missing answer an error and are not charged.
Can an agent pay per page without a key?
Yes. An agent with its own wallet can pay for each page as it reads it; the For AI agents section below has the details.
How do I read a whole site?
Read its sitemap with the sitemap tool (up to 100 URLs a call, with an offset to page through), then read each URL.
Is there a subscription?
There is a monthly plan and there are one-off credit packs. Both give one API key; a plan adds credits each month and unused credits stay.
Guides
For AI agents
The same tool is on the MCP server https://glyphcall.com/mcp (the first 25 successful calls are free), and an agent with a wallet can pay per call in USDC: the tool URL answers 402 Payment Required with x402 and MPP terms. Reference: tool page · OpenAPI · docs for agents.