46 Web Scraping Prompts for ChatGPT & Claude
Built-in browsing in Claude and ChatGPT struggles with JavaScript, pagination, and bot protection. These web scraping prompts hand the job to Context, which renders the page, returns clean Markdown or JSON in the shape you ask for, and can map or crawl a whole site, so you get data instead of code to debug.
Web Scraping prompts
Product page to JSON
Scrape [product URL] and return JSON with name, brand, price, currency, availability, variants, rating, review count, and image URLs. Use null for any field that isn’t on the page.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Whole category to CSV
Scrape every product in the category at [category URL], following pagination. Output a CSV with name, URL, price, and stock status, and tell me how many pages you visited.
Uses: Scrape, Map URLs, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Price comparison across stores
Find [product] on [store 1], [store 2], and [store 3]. Scrape the current price, shipping cost, and delivery estimate from each, and rank them by delivered price.
Uses: Web Search, Scrape. Install the Context.dev MCP if you haven’t already.
Every table to a spreadsheet
Extract every table on [URL] into CSV. Keep headers, merge split rows, and convert numbers and dates to consistent formats.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Docs site to Markdown for RAG
Crawl the documentation at [docs URL], up to [N] pages. Return each page as clean Markdown with its URL and title, skipping navigation, footers, and cookie banners, then split it at headings into chunks of about [500] tokens as JSONL with URL, heading path, and text.
Uses: Crawl & Batches, Map URLs. Install the Context.dev MCP if you haven’t already.
Every URL on a site
List every URL on [domain]. Group them by section (blog, docs, product, legal), count each group, and flag URLs that look like duplicates or test pages.
Uses: Map URLs. Install the Context.dev MCP if you haven’t already.
Job listings dataset
Scrape all open roles from [careers page URL], including the job board pages it links to. Return title, team, location, remote policy, salary if listed, and posting URL.
Uses: Scrape, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Real estate listings
Scrape the listings on [listing search URL]. Return address, price, beds, baths, square footage, days on market, and listing URL, then sort by price per square foot.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Reviews ready for analysis
Scrape the reviews for [product] at [review page URL], following pagination up to [N] reviews. Return rating, date, title, and text, then summarize the top complaints.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Screenshot a list of pages
Take full-page screenshots of these URLs at desktop width: [list]. Tell me which pages failed to load or showed a cookie wall instead of content.
Uses: Scrape, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Pull every image from a page
Get every image on [URL] with its source URL, alt text, and dimensions, dropping visual duplicates. Flag images missing alt text and point out the likely product hero shots.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Scrape behind a “Load more” button
Open [URL] and click “[button text, such as Load more]” until all items show, up to five clicks. Then extract [fields] for every item.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Infinite scroll pages
Scroll [URL] to load more results, then extract the [items] with [fields]. Tell me how many items loaded and whether more were still coming.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Same page from another country
Scrape [URL] as a visitor from [country] and as a visitor from the US. Compare prices, currency, available products, and any content that differs.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Directory to a lead table
Scrape the directory at [URL] into a table with name, category, location, website, and description. Open each listing’s detail page when fields are missing.
Uses: Scrape, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Event listings
Scrape upcoming events from [venue or events URL]. Return name, date, time, price range, venue, and ticket link in chronological order.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Restaurant menus
Scrape the menus at [restaurant URLs]. Return dishes with prices, dietary labels, and sections, and compare the average entree price across restaurants.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Structured data extractor
Pull all schema.org structured data (JSON-LD and microdata) from [URL] and return it as clean JSON grouped by type, such as Product, Offer, Review, and Organization. Point out any value that disagrees with what the page visibly shows, like a different price.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Batch scrape a URL list
Scrape these [N] URLs: [list or file]. For each, return title, meta description, H1, word count, and main text as Markdown, and list any failures with the reason.
Uses: Crawl & Batches, Scrape. Install the Context.dev MCP if you haven’t already.
Changelog to timeline
Scrape the changelog at [URL] and turn it into a dated timeline of releases with a one-line summary each. Highlight everything shipped in the last 90 days.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Contact pages at scale
For these company websites [domains], find the contact or about page and extract the public business email, phone, address, and social links. Leave a field blank rather than guessing.
Uses: Map URLs, Scrape. Install the Context.dev MCP if you haven’t already.
Watch a page for changes
Monitor [URL] every [day or hour] and alert me when [specific element, such as price, stock status, or a section] changes. Ignore edits elsewhere on the page, and show the before and after.
Uses: Monitors. Install the Context.dev MCP if you haven’t already.
Every PDF on a site
Find every PDF linked on [domain], such as reports, spec sheets, and whitepapers. List title, URL, and the page that links to it, then summarize the five most recent.
Uses: Map URLs, Crawl & Batches, Parse Files. Install the Context.dev MCP if you haven’t already.
Diff two versions of a page
Scrape [URL A] and [URL B], for example staging and production. Compare text, headings, links, and images, and list every difference.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Turn it into a reusable scraper
Scrape [URL] for [fields] and show me the result. Then write a [Python or TypeScript] script with the Context SDK that extracts the same JSON schema from any page like it.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Extract into my JSON schema
Scrape [URL] and return JSON that matches this schema exactly: [JSON schema or list of fields with types]. Use null for anything the page doesn’t show, keep numbers as numbers, and list any field you had to infer rather than read directly.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Entire store catalog to JSON
Map every product URL on [store domain], then scrape each product page into JSON with name, SKU, brand, price, sale price, currency, variants, stock status, and image URLs. Give me the full catalog as a CSV, the product count, and the URLs that failed.
Uses: Map URLs, Crawl & Batches, Scrape. Install the Context.dev MCP if you haven’t already.
Thousands of URLs to one dataset
Run a batch job over these [N, up to 100,000] URLs: [file or sitemap URL]. Extract [fields] from each page into one CSV with the source URL in the first column, then list every failed URL with the reason so I can retry just those.
Uses: Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Price drop and restock alerts
Monitor these product pages daily: [product URLs]. Track price and stock status, and alert me only when a price drops below [target price] or an out-of-stock item comes back. Send each alert to [Slack or webhook URL] with the old value, new value, and link.
Uses: Monitors, Scrape. Install the Context.dev MCP if you haven’t already.
Scrape the top search results
Search for “[query]” and scrape the top [20] result pages. Pull [fields, such as price, plan names, or publish date] from each into one table with the source URL, and mark pages where a field isn’t present instead of guessing.
Uses: Web Search, Scrape. Install the Context.dev MCP if you haven’t already.
Google Maps results to a CSV
Scrape the Google Maps results for “[business type] in [city]” into a CSV with name, category, rating, review count, address, phone, website, and Maps link. Sort by review count and flag places with no website.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Alert me to new listings
Monitor [search results URL for rentals, jobs, or marketplace items] every [few hours] and alert me only when new listings appear, ignoring reordering and ads. For each new listing, give [fields, such as title, price, and location] and the link.
Uses: Monitors. Install the Context.dev MCP if you haven’t already.
Blog to Markdown files
Find every post on [blog URL] and convert each one to a Markdown file with front matter for title, date, author, tags, description, and original URL. Keep headings, images, and code blocks, and drop share buttons, comments, and related-post widgets.
Uses: Map URLs, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Q&A dataset from a docs site
Crawl [docs or help center URL] and write [N] question-and-answer pairs grounded in the pages, as JSONL with question, answer, source URL, and the supporting quote. Cover every major section and skip anything the text doesn’t actually answer.
Uses: Crawl & Batches, Map URLs. Install the Context.dev MCP if you haven’t already.
Content inventory before a migration
Map every URL on [domain] and scrape each page’s title, H1, word count, page type, and last updated date into a content inventory spreadsheet. Flag pages with forms, embedded videos, or PDFs that a move to [new platform] could break.
Uses: Map URLs, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Migration parity check
Compare [old site] with [new site] page by page, matching URLs by path or with this mapping: [old and new URL pairs]. For each pair, compare title, H1, word count, images, and links, and list pages where content went missing or the new URL is broken.
Uses: Map URLs, Crawl & Batches, Scrape. Install the Context.dev MCP if you haven’t already.
Use a site’s own search box
Open [URL], type “[search term]” into the site’s search box, and submit. Extract every result with [fields], and tell me how many results the site says it found compared with how many you captured.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Content hidden behind tabs
Open [URL] and click through each [tab, accordion, or toggle, such as monthly and annual billing], up to five clicks. Return the content each one reveals as Markdown labeled with the tab name, and note anything that never loaded.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
App store reviews by theme
Scrape the newest reviews for [app] from its App Store and Google Play pages. Group praise and complaints into themes with counts and two short quotes each, and list the bugs users report most with the date they first appeared.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
API docs to an OpenAPI spec
Crawl the API reference at [docs URL] and draft an OpenAPI 3.1 spec in YAML with every endpoint, method, parameter, request body, and response field it documents. List the endpoints whose details the docs leave unclear.
Uses: Crawl & Batches, Map URLs, Scrape. Install the Context.dev MCP if you haven’t already.
Mobile vs desktop screenshots
Take full-page screenshots of [URL] at mobile width (390 px) and desktop width (1440 px). Compare them and list mobile layout problems such as cut-off text, overlapping elements, oversized images, or a hidden call to action.
Uses: Scrape. Install the Context.dev MCP if you haven’t already.
Store locations to a CSV
Scrape every location from the store locator at [URL] into a CSV with store name, address, city, region, postal code, phone, and opening hours. Open individual location pages when the list leaves fields out, and give totals by [country or state].
Uses: Scrape, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Wikipedia list to a dataset
Scrape the table on [Wikipedia list URL] into CSV, then open each row’s linked article and add [fields, such as founding year or headquarters] from its infobox. Mark rows with no linked article or a missing value.
Uses: Scrape, Crawl & Batches. Install the Context.dev MCP if you haven’t already.
Minimum advertised price check
Search online retailers for [product name or SKU] and scrape the advertised price on each listing. List every seller below our minimum advertised price of [MAP price] with the seller name, URL, price, and date checked.
Uses: Web Search, Scrape. Install the Context.dev MCP if you haven’t already.
Terms and policy change tracker
Monitor the terms of service, privacy policy, and data processing pages of [vendors or tools we use] weekly. When the wording changes, quote the old and new text, explain the change in plain English, and flag anything about data use, fees, or liability.
Uses: Monitors. Install the Context.dev MCP if you haven’t already.
Write an llms.txt for a site
Map [domain] and read its most important pages. Write an llms.txt file with a one-paragraph summary of the site, then grouped sections of links (docs, product, pricing, policies) with a one-line description each, capped at [N] links.
Uses: Map URLs, Scrape. Install the Context.dev MCP if you haven’t already.
How to use these prompts
Install the Context.dev MCP
Add the Context.dev MCP server to Claude, Cursor, or another agent, or install the ChatGPT app. This gives the assistant live web search, scraping, and data tools.
Pick a prompt
Copy it, or open it straight in Claude or ChatGPT. Every copied prompt tells the assistant to use Context.dev, links the MCP install page, and asks it to cite sources.
Fill in the brackets
Replace every [bracket] with your details, such as a company domain, a competitor, or a file, then run it.
Context APIs behind these prompts
The tools your assistant calls when it runs these prompts.
Web Scraping prompt FAQ
Can ChatGPT scrape a website?
Not reliably on its own. Connected to Context, it can scrape any public page into Markdown, HTML, screenshots, images, or JSON that matches a schema you describe, and it can map or crawl an entire site.
Do I need to write code to use these prompts?
No. Describe the data you want in plain language and the assistant calls the scraping tools for you. If you want a reusable script afterwards, ask it to write one with the Context SDK.
Which sites can I scrape?
Public pages, including JavaScript-heavy sites. Respect each site’s terms and rate limits, and don’t collect personal data you have no lawful basis to process.
How do I install the Context.dev MCP in Claude or ChatGPT?
Add the Context.dev MCP server to Claude, Cursor, or any MCP client with the MCP setup guide, or install the Context app in ChatGPT for web search and page reading. Then paste a prompt and the assistant picks the right Context tools.
Do these prompts work without Context?
They run, but without live web tools the assistant answers from its training data and can’t open the pages, files, or searches the prompt asks for. Expect older information and more guesses.
More prompt categories
- GTM promptsICP research, positioning, pricing, launches, and market maps.65 prompts
- Coding promptsBuild scrapers, agent tools, and brand features into your app.53 prompts
- Lead Generation promptsProspect lists, decision makers, buying signals, and enrichment.48 prompts
- Cold Email promptsResearched first lines, sequences, follow-ups, and reply handling.36 prompts
- Web Search promptsDeep research, fact-checking, sourcing, and comparisons with citations.33 prompts
- File Parsing promptsPDFs, contracts, invoices, spreadsheets, slides, and scans.28 prompts
- News promptsCompany news, competitor alerts, briefings, and media monitoring.31 prompts
- Ad Copy promptsAd copy, hooks, audiences, landing pages, and competitor messaging.27 prompts
- Creative & Design promptsOn-brand image prompts, creative briefs, mockups, and brand kits.26 prompts
- Competitor Analysis promptsPricing, features, reviews, hiring, and change tracking.41 prompts
- SEO promptsSite audits, content briefs, search results research, and internal links.38 prompts
- Recruiting promptsCandidate sourcing, research, pay benchmarks, and outreach.29 prompts
Start building with live web context.
Scrape, search, and research the web with one API. Bring the results straight into your product.