Web Scraping API, LLM-ready out of the box.
Get Markdown, HTML, JSON or plain text from any webpage, including JavaScript ones. Rendering, CAPTCHA solving, and bot detection are handled for you, starting at 1 credit a request.
- 500 free credits every month
- No card to start
- Failed requests cost nothing
POST /webscraping
{
"url": "stripe.com",
"format": [
"markdown-llm"
]
}
The six problems you don't have to solve.
Each is a system you would otherwise have to build, run, and maintain. The first five are included with every request at 1 credit.
JavaScript rendering
Pages are fetched without a browser first, then rendered in headless Chrome only when needed. React, Vue, and Angular applications return their rendered content, while static pages stay fast.
renderJSCAPTCHA solving
Common CAPTCHA challenges are handled in the request path. There's no separate solver to integrate and no extra charge.
includedAnti-bot bypass
Browser fingerprints are managed on our side to handle common anti-bot systems. Stealth mode reduces browser automation signals for stricter sites.
stealth: trueSeven output formats
Choose from Markdown, HTML, or plain text, each with an LLM variant, plus structured JSON. Request multiple formats in a single call.
format: []Structured extraction
Extract structured fields with CSS or XPath schemas, or use ready-made product and contact templates to get typed JSON.
extractionModeGeo-targeted proxies
Route requests through a proxy in any of 194 countries to see localized pricing, content, and availability. Geo-targeting is opt-in and adds 4 credits when used.
proxyCountrySee the page the way locals see it.
Some sites block foreign traffic, and many show different prices and content by country. Add a country code and the request goes through a proxy, so you get the local page and don't get blocked.
proxyCountryroutes a request through any of 194 countriesproxyMode: "auto"tries direct first and uses a proxy only if the site blocks you- Adds 4 credits only on requests that use a proxy
- 0
- Countries
- Country
- Level of geo-targeting
- Opt-in
- Only when you need
The parameters you'll use most.
See what each parameter does, what it costs, and when to use it without digging through the API reference.
| Parameter | Type | What it does | Credits |
|---|---|---|---|
| url | string | The page to fetch. Any public URL. | — |
| format | string[] | markdown, markdown-llm, html, html-llm, text, text-llm or json. Ask for several at once. | included |
| renderJS | boolean | Rendering is automatic. Set true or false to force headless Chrome on or off. | included |
| stealth | boolean | Removes webdriver signals and patches navigator properties for sites with stricter bot checks. | included |
| waitTime | number | Seconds to wait after load, for lazy-loaded content. | included |
| device | string | desktop or mobile. | included |
| extractionMode | string | With format json: cssSchema, xpathSchema, or template with product or contact. | included |
| fileOutput | boolean | Return a link to a file with the result instead of the content inline. Handy for large pages. Defaults to false. | included |
| proxyCountry | string | Route the request through a proxy in this country, e.g. "de". | +4 |
| proxyMode | string | auto tries without a proxy first and uses one only if the site blocks the request. | +4 if used |
| aiPrompt | object | Ask AI to extract or analyze the page. Runs on the page's Markdown. | +6 |
These are the parameters you'll use most. The endpoint reference has the complete parameter list, response schema, and error codes.
Endpoint referenceWhat people actually use it for.
Clean Markdown straight into a vector store
Feed webpages into a retrieval pipeline without building your own extraction layer. Our LLM-ready format removes navigation, footers, cookie banners, and other page clutter, leaving the content you actually want to embed.
- Headings are preserved for meaningful chunk boundaries
- Navigation, footers and cookie banners removed
- Multiple formats from a single request when you need both
POST /webscraping
{
"url": "https://techcrunch.com/category/artificial-intelligence/",
"format": ["markdown-llm"]
}Your first scraping, in a few lines.
# pip install geekflare-api
from geekflare_api.client import GeekflareClient
from geekflare_api.models import WebScrapeDto
with GeekflareClient(api_key="<api-key>") as client:
result = client.web_scrape(
WebScrapeDto(
url="https://example.com",
format=["markdown"]
)
)
print(result)One API. Every stack.
The same key works from an agent, a workflow builder, an SDK or your own service. Pick the lane you already live in.
MCP server
One remote endpoint exposes Geekflare's tools to any MCP client. Your agent discovers them itself: no glue code, no per-tool wiring.
Claude
Cursor
Codexany MCP clientLLM & agent pipelines
LLM-ready Markdown straight into a retrieval pipeline, with the boilerplate stripped so you spend context on content, not navigation.
Workflow builders
Run any endpoint as a step in a visual scenario. Useful when the person who needs the data does not write code.
SDKs & REST
First-party typed SDKs for Python and Node. Everything else talks plain REST, from any language.
PythonNode.jsPostmanOne credit a page. Here is what that buys.
Credits are shared with every other Geekflare endpoint. Failed requests cost nothing.
Growth covers up to ~100,000 pages a month (100K credits), about $0.69 per 1,000 pages. Or $58/mo billed yearly.
- 500 credits / month
- 1 team member
- 7 days log retention
- 1 request per second
- 10K credits / month
- 3 team members
- 30 days log retention
- 5 requests per second
- 100K credits / month
- 5 team members
- 30 days log retention
- 10 requests per second
- 1M credits / month
- 25 team members
- 90 days log retention
- 25 requests per second
Teams that stopped maintaining scrapers.
“Found Geekflare API to get markdown from URL for my AI agents. It is fast and cheaper and works on almost every website.”
Ram DasiArchitect, PA Consulting“After trying many scraping services, I selected Geekflare to scrape public directories. Mainly for two reasons - it is cheaper and fast.”
“Documentation was easy to follow. We were scraping dynamic pages within hours. Very reliable service.”
Before you write the integration.
Everything else is in the endpoint docs, and our support team replies to every email.
Open the endpoint docsaiPrompt adds 6. Failed requests cost nothing.renderJS to force either way.proxyCountry with a country code, for example "de", to route the request through a proxy in that country and get localized pricing and content. Proxies cover 194 countries at country level and add 4 credits per request.markdown-llm for LLM and RAG pipelines, json when you want typed fields, html when you have your own parser downstream, and text for search indexing or NLP. You can request more than one format in a single call.format: ["json"], use a CSS or XPath schema to pick fields, or set extractionMode to template with product or contact for ready-made extraction. For free-form questions about a page, aiPrompt runs AI over its Markdown.stealth: true removes webdriver signals.robots.txt and avoid collecting personal information.500 pages, no card, one curl away.
The free tier renews every month. Scrape one of the pages and see what comes back.