Priced by bandwidth, not requests.
Retrieval is billed by bandwidth. Search uses credits; structured extraction uses both. Charges depend on the endpoint and options.
Web scraping. Without the upkeep.
Turn public pages into Markdown or structured data for your agents and pipelines. ByteKit handles stealth, fetching, browser rendering, and extraction.
npm install @hunt-labs/bytekit-sdk pip install bytekit-sdk50% less on cache hits.
Scrape cache hits halve the bandwidth charge. Choose your cache window, or fetch fresh content.
Unused bandwidth rolls over.
Carry unused monthly bandwidth forward with no cap while you keep your plan or upgrade.
Integrate in minutes.
Get the web data your application needs without building the retrieval infrastructure.
Retrieval that fits your workflow.
Browser rendering, content conversion, caching, and scheduled checks—without running the infrastructure yourself.
Rendering when needed.
ByteKit can fall back to a browser when an HTTP fetch isn’t enough. Set wait conditions for content that loads later.
Choose your output.
Request Markdown, raw HTML, or structured fields. Capture screenshots or discover URLs through sitemaps.
Search, then retrieve.
Find URLs with search, then retrieve the pages you choose. Already have a URL? Start with retrieval.
Checks on your schedule.
Configure a monitor and webhook. ByteKit re-fetches the page and notifies your application when content changes.
Caching for lower latency & cost.
Scrape cache hits cost 50% less in bandwidth charges. Choose your cache window, or turn caching off for a fresh fetch.
Content with page metadata.
Retrieve content with available page metadata, such as title, language, and description, for your application to use.
Ways to put ByteKit to work.
Give agents context, build datasets, or track changes on public pages.
Agent web access
Let agents retrieve public pages as Markdown, with search available when needed.
RAG and corpus ingestion
Retrieve Markdown and page metadata for your chunking, embedding, and indexing pipeline.
Training and evaluation
Capture pages on a schedule for your training and evaluation datasets.
Competitor monitoring
Monitor public pricing and product pages. Receive a webhook when page content changes.
Product catalog enrichment
Extract prices, specifications, and availability from public product pages into fields you define.
Sales and account research
Extract industry, product, and location details from public company pages for account research.
Own the workflow. Ditch the retrieval stack.
Keep your application logic. Hand off anti-bot handling, browser infrastructure, extraction, and monitoring.
| Responsibility | Building it yourself | With ByteKit |
|---|---|---|
| Anti-bot handling | Maintain compatibility as anti-bot protections change. | ByteKit maintains anti-bot handling. |
| Browser infrastructure | Operate browsers and manage rendering, failures, and retries. | A simple API request. We choose the retrieval strategy. |
| Extraction & formatting | Maintain parsing and conversion logic. | Request Markdown or define fields with a JSON schema. |
| Keeping data current | Schedule checks, detect changes, and deliver notifications. | Configure a monitor and webhook. |
Public web data. Clear boundaries.
Use ByteKit for public pages you’re authorized to access.
Public pages
Login workflows and paywalled content aren’t supported.
Restricted destinations
We block .gov domains and selected financial, healthcare, ticketing, and social media sites.
No ticketing abuse
No ticket-purchase bots, queue jumping, reservation bots, or inventory hoarding.
No personal surveillance
No doxxing, stalking, or unauthorized tracking of individuals.
No fraud or harmful use
Phishing, impersonation, scams, and malware-related use are prohibited.
Respect access rights
Technical access isn’t permission. Respect restrictions and requests to stop.
Try ByteKit for free.
New accounts receive a one-time allowance of 100 MB and 100 credits.