Reworkd

What it does

Reworkd takes a list of target web pages or URLs and returns structured data extracted from them. Its AI agents read each page, generate extraction code for the fields you request, run that code, validate the results, and deliver the data. The pipeline also handles pagination, infinite scroll, dynamic content, retries, rate limiting, and proxies.

Teams use Reworkd to collect and monitor web data on a recurring basis without writing and maintaining their own scrapers. Its extraction code is generated for each requested field, and its scrapers detect and repair extraction failures when page layouts change. An analytics dashboard shows what is being extracted and which runs succeed or fail. The platform is also used to download large volumes of regulation PDFs.

Reworkd fits teams that need ongoing web data collection without in-house scraping engineering. You sign up at app.reworkd.ai, and the documentation is linked from the website. Plans include a free Hobby tier and a $99 per month Pro tier, with separate usage charges for proxies, compute, and captcha solving. Concurrent browser limits and data retention periods depend on the plan.

What it can do

• Generates extraction code from web pages to pull requested data
• Runs extractors and validates extracted results
• Self-healing scrapers that detect content changes and repair failures
• Handles pagination, infinite scroll, and dynamic content
• Scheduled jobs for recurring extraction
• Analytics dashboard showing extraction status and changes

Example workflows

Recurring product catalog extraction

Input: a list of retailer product-listing URLs and the fields to capture (name, price, availability). Steps: the agent reads each page, generates extraction code for those fields, runs it on a schedule, validates the output, and repairs extractors if a page layout changes. Output: a structured data set refreshed on each run, with a dashboard showing success and failure counts.

Regulatory document collection

Input: a set of government web portals listing regulation documents. Steps: the agent scans the portals, handles pagination, generates extraction code to locate and download PDFs, and validates the files. Output: a collection of downloaded PDF documents ready for downstream processing.

Before you choose

• Usage-based infrastructure charges apply on top of the plan price: standard proxies at $0.125 per GB, premium proxies at $8 per GB, compute at $0.10 per hour, and antibot solving at $5 per 1,000 solved.
• Hobby plan data retention is 30 days and Pro is 90 days; Enterprise retention and concurrency are custom.
• The site states that generated code is used rather than model predictions, but output quality still depends on the target site and should be validated.

Listed under

• Data
• Data extraction
• AI agents
• Developers
• No-code
• Proxies
• Web scraping

Get started

Create an account at app.reworkd.ai; the Hobby plan is listed at $0 per month.

1. Create an account at app.reworkd.ai.
2. Review the documentation at reworkd.ai/docs to learn how to define extraction targets.
3. Choose a plan on the pricing page and configure concurrent browser limits, data retention, and scheduled jobs.

Pricing page lists plan prices and usage-based infrastructure rates. Billing cadence and seat basis beyond the displayed monthly prices are not stated. Check the pricing page for current terms.

Source: reworkd.ai/pricing, checked 29 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Reworkd ↗www.reworkd.ai
• Reworkd ↗www.reworkd.ai

Company website

reworkd.ai: product, pricing and policy pages

Hacker News

Launch posts and discussion

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.