# WebReaderAPI FAQ > WebReaderAPI FAQ: 30 answers on the web reader API: price per page, JavaScript and PDFs, crawls, robots.txt, blocked sites, KYC, crypto payments and AI agents. - URL: https://webreaderapi.com/faq/ - Last updated: 2026-10-22 - Publisher: WebReaderAPI (https://webreaderapi.com) Answers to what developers, AI builders and site owners ask most about turning web pages into Markdown with WebReaderAPI. Each answer starts with the answer. **Quick answer — What do people ask most about WebReaderAPI?** Mostly the price per page, how JavaScript pages and PDFs are counted, and what happens when a site refuses robots. In short: **WebReaderAPI** charges one credit per page read, two with a browser, from $0.07/1K pages as of October 2026; refused or failed pages cost nothing; robots.txt is always respected. No KYC; pay in crypto. ## Price and credits ### How much does WebReaderAPI cost? Prepaid packs cost $3.75 for 10,000 pages, $21.75 for 100,000 pages and $74.25 for 1,000,000 pages, as of October 2026. The rate falls to $0.07/1K pages on the 5,000,000 pages pack. The [pricing page](https://webreaderapi.com/pricing/) lists every pack. ### What does one credit buy? One page read without a browser. A page that needs JavaScript rendering uses two credits, and reading from a chosen country adds one. The response tells you what was used in `credits_used`, so you can track cost per page in your own logs. ### Do credits expire? No. Prepaid credits never expire: unused pages stay on your account until you use them. A pack bought for a one-off crawl keeps its unused pages for the next project. ### Am I charged when a page cannot be read? No. Errors use no credit, whether the cause is our side, the network or the website. That includes pages refused by robots.txt, by an `ai-input=no` signal, or by the site's own answer. ### Is there a free tier? No. There is no free tier: the smallest pack is the way to test, paid from the minimum top-up, and the amount added only to reach that minimum can be sent back on request within 30 days, in USDT or USDC, while unspent. ### What is the smallest amount I can spend? The minimum top-up is $25. The smallest pack, 10,000 pages, costs $3.75; the rest of the top-up stays on your balance for later packs or can be asked back within 30 days if unspent. ## What the API reads and returns ### Can WebReaderAPI read pages built with JavaScript? Yes. JavaScript rendering in a headless browser when the page needs it (render=auto), or always, or never. With `render=auto`, a browser is used only when the raw HTML holds too little text, so simple pages stay at one credit. ### Can it convert PDF files? Yes. PDF files up to 20 MB and 100 pages are converted to Markdown, at 1 credit per started block of 10 pages. Send the PDF's URL to the same `/read` endpoint; the text comes back as Markdown. ### Can I read a whole website? Yes, with a crawl job. Crawl one website per job, up to 10,000 pages, following its sitemap, its robots.txt and its Crawl-delay. You can restrict it to paths such as `/docs/*`; each page read uses credits like a single request. ### How many URLs can I send at once? Up to 100 URLs per batch job. A batch runs as a job: you send the list, receive a `job_id`, and collect every result when the job is done. ### How do I keep a page within my model's context window? Set `max_chars`. The Markdown is cut at that length, from 1,000 to 500,000 characters. Combined with `main_content=true`, this keeps long pages within a predictable token budget. ### Does the Markdown keep tables, code blocks and links? Yes. Headings, lists, tables, code blocks and links keep their Markdown form. Links are inline; add `include_links=true` to also get them as a separate list, and `include_images=true` to keep images. ### What does "main content" mean? The part of the page a reader came for: the article, the documentation or the product description. With `main_content=true`, the default, menus, cookie banners, ads, sidebars and footers are removed. Set it to `false` to convert the whole page. ### Can I read a page as it appears in another country? Yes, with the `country` parameter and a two-letter country code. Choosing the country a page is read from adds 10 credits, because the page is then read through a residential connection in that country. Some sites change prices, language or content by country. ## robots.txt and refused pages ### Does WebReaderAPI respect robots.txt? Always. We respect robots.txt: our crawler identifies itself as WebReaderAPI, follows the rules written for it or for all robots, and does not read a page that a site disallows. Refused pages are not charged. The [bot page](https://webreaderapi.com/bot/) shows the user agent and how site owners can allow or refuse it. ### What happens with sites protected by Cloudflare or another bot filter? If the site lets our crawler through, the page is read normally. If it refuses or answers with a challenge page, we return that answer as `target_refused` and charge nothing. We never try to get past a challenge. ### Can WebReaderAPI read pages behind a login or a paywall? No, never, even with credentials you provide. Content that requires an account or a subscription stays out of scope, and asking for it breaks the [Acceptable Use Policy](https://webreaderapi.com/acceptable-use/). ### What is the ai-input=no signal? A line some sites add to robots.txt, from the Content Signals Policy, to refuse the use of their pages as input for AI answers. If a site declares Content-Signal: ai-input=no in its robots.txt, we do not read its pages for you; the API returns `blocked_by_site_policy`, free of charge. ### Is reading web pages for AI legal? It depends on the content, the country and what you do with the text. Respecting robots.txt and staying away from personal data and paywalls lowers the risk. Our [overview of the rules](https://webreaderapi.com/guides/is-reading-web-pages-for-ai-legal/) summarises the main points; it is not legal advice. ## Account, payment and privacy ### Do I need to verify my identity? No. WebReaderAPI asks for no ID, no selfie, no company documents and no card. Account-number login; email is optional. The [no-KYC page](https://webreaderapi.com/no-kyc-web-reader-api/) explains what still applies. ### Which cryptocurrencies can I pay with? BTC, USDT, USDC, ETH, SOL, LTC. USDT works on TRC-20 or ERC-20, USDC on ERC-20. Send the exact amount shown, click "I've sent the payment", and the top-up is credited once it is confirmed. ### Can I get a refund? Yes, to your balance. A pack used at less than 20% goes back within 72 hours of purchase, and a pack that never worked is credited in full. The [refund policy](https://webreaderapi.com/refunds/) has the exact rule and three examples. ### How do API keys work? Create several API keys, for example one per agent or project, and revoke any of them from your dashboard. Send your API key in the Authorization header as a Bearer token, or in the api_key query parameter. One key per application or agent makes it easy to revoke one without stopping the others. ### What do you log about my requests? The host name of each page, never the full URL or its content, plus the time, status and credits, for 30 days. We keep no copy of the pages you read. The [privacy policy](https://webreaderapi.com/privacy/) has the details. ## Developers and AI agents ### Is there code for Python and Node.js? Yes. The [Python](https://webreaderapi.com/docs/python/) and [Node.js](https://webreaderapi.com/docs/node/) guides show single reads, batches, crawls and retries, and the [quick start](https://webreaderapi.com/docs/) has a curl example. No SDK package is needed: the API is plain HTTPS with JSON. ### Can Claude, Cursor or another MCP client use WebReaderAPI? Yes, through a small MCP server you run yourself, documented on the [MCP page](https://webreaderapi.com/docs/mcp/). It exposes a tool that takes a URL and returns the page as Markdown. ### Does it work with LangChain and LlamaIndex? Yes. The [LangChain and LlamaIndex page](https://webreaderapi.com/docs/langchain/) shows a document loader that turns URLs into documents with their metadata, ready for chunking and embedding. ### How many pages can I read at the same time? Every account has a limit on how many pages it reads at once; your dashboard shows it, and support can raise it. Crawls also follow each site's `Crawl-delay`, so a large crawl of one domain may take longer than its page count suggests. ### Who runs WebReaderAPI? SerpGoat LLC. See the [about page](https://webreaderapi.com/about/) for how the service works and how to reach a person. Email support answers within 12 hours, every day of the week, including help with crawls and integrations. ### How fast is a read? It depends mostly on the website and on whether a browser is needed. We do not publish a speed figure until we have measured one over a public sample; when we do, it will appear with its method and date on the [status page](https://webreaderapi.com/status/). --- WebReaderAPI: WebReaderAPI is a web reader API for AI agents that turns any public web page into clean Markdown, paid in crypto with no ID upload. Prepaid credits from $0.07/1K pages never expire, and robots.txt is respected. No KYC — no ID, no selfie, no documents. Pay in crypto.