Documentation · Collectors · MCP server · Blog · Free API key
Web data infrastructure: six REST endpoints, 31 ready-made scrapers, an MCP server for AI agents, and six proxy networks underneath all of it. One Bearer key, one JSON envelope, and you are only billed for calls that succeed.
This repository is the index. Every example linked below is a real, runnable repo.
- Quick start
- The Data APIs
- Collectors
- Proxy networks
- For AI agents
- Examples by use case
- Examples by language
- Pricing model
export QD_API_KEY=qd_live_your_key_here
curl https://api.quanticdata.io/v1/scrape \
-H "Authorization: Bearer $QD_API_KEY" \
-H "Content-Type: application/json" \
-d '{"url": "https://example.com", "format": "markdown"}'Same envelope on every endpoint: type, message, payload. Failures answer
{"type": "error", "message": "…"} and cost nothing.
Eight languages, side by side: quanticdata-api-quickstart
| Endpoint | What it does | Price | Examples |
|---|---|---|---|
POST /v1/scrape |
one page → Markdown, HTML, text, CSS or AI extraction, screenshots | $0.0002 | web-scraping-api-examples |
POST /v1/serp |
4 engines, 18 verticals, parsed to JSON | from $0.0005 | serp-api-python-client |
POST /v1/map |
every URL of a site, sitemaps + links, flat price | $0.0005 | crawl-and-map-api-examples |
POST /v1/crawl |
async BFS crawl → Markdown per page | $0.0003/page | crawl-and-map-api-examples |
POST /v1/batch |
up to 1,000 known URLs, one job id | $0.0002/URL | batch-url-scraping-jobs |
POST /v1/seo-audit |
no-JS view vs rendered view, diffed | $0.0012 | seo-audit-api-examples |
Full reference: quanticdata.io/docs
Thirty-one versioned scrapers you call with a semantic input — a keyword and a city, an ASIN, a domain, a handle — instead of a URL and a set of selectors. Published input and output schemas, an hourly health probe, and a price per delivered row.
Start here: collectors-api-examples · catalogue at quanticdata.io/collectors
| Area | Collectors | Examples |
|---|---|---|
| Search & SEO | web_search search_images search_videos keyword_ideas |
keyword-research-api-examples |
| Local | google_maps_places place_reviews |
google-maps-places-api-examples · google-reviews-api-examples |
| E-commerce | amazon_search amazon_product ebay_search aliexpress_search google_shopping product_offers |
amazon-product-api-examples · competitor-price-monitoring-api |
| Jobs | linkedin_jobs indeed_jobs google_jobs |
linkedin-jobs-api-examples · indeed-jobs-api-python |
| Social & video | youtube_search youtube_channel instagram_profile tiktok_profile tiktok_video reddit_posts |
youtube-scraper-api-examples · instagram-tiktok-profile-api |
| Companies & leads | company_profile site_contacts local_business_leads linkedin_company linkedin_profile |
lead-scraper-api-examples · company-data-enrichment |
| Apps | app_store_apps google_play_apps |
app-store-data-examples |
| Real estate | zillow_search |
zillow-listings-api-examples |
| News | google_news |
reddit-news-monitoring |
| Travel | hotels |
hotels-and-travel-data-api |
| Network | Gateway | Page |
|---|---|---|
| Residential Premium | pr.quanticdata.io:7777 |
residential-proxies |
| Residential / rotating | resi.quanticdata.io:8080 · rotate.quanticdata.io:8000 |
cheap-residential-proxies · rotating-proxies |
| SOCKS5 | resi.quanticdata.io:1080 |
socks5-proxies |
| Mobile | mb.quanticdata.io:7777 |
mobile-proxies |
| Datacenter | dc.quanticdata.io:7777 |
datacenter-proxies |
| ISP | isp.quanticdata.io:8000 |
isp-proxies |
| IPv6 | v6.quanticdata.io:7777 |
ipv6-proxies |
Targeting is expressed in the username — USER-country-de-city-berlin-session-a91f-sessTime-30.
Full syntax and eight-language examples:
quanticdata-proxy-quickstart
Also: mobile-proxy-examples · ipv6-proxy-examples · sneaker-proxy-setup
Everything above is exposed as MCP tools, so an agent can search, read, map, crawl and run collectors on its own.
- quanticdata-mcp-server — a single-file, zero-dependency MCP server for Claude Code, Claude Desktop, Cursor and anything else that speaks the protocol
- web-data-for-ai-agents — the same tools wired into Anthropic and OpenAI tool-calling loops
- rag-corpus-builder — sites → chunked, de-duplicated, cited JSONL
- ai-extraction-examples — CSS selectors vs LLM extraction, and the hybrid that survives a redesign
- browser-ai-agent-examples — a goal-driven agent loop on the live APIs
Product pages: Web Data API for AI · MCP server · AI web scraping · Browser AI
| I want to… | Repo |
|---|---|
| watch competitor prices | competitor-price-monitoring-api |
| build a lead list with contacts | lead-scraper-api-examples |
| enrich a list of domains | company-data-enrichment |
| size a market from live data | market-research-datasets |
| track brand mentions | reddit-news-monitoring |
| plan content from real demand | keyword-research-api-examples |
| check a site is indexable without JS | seo-audit-api-examples |
| feed a RAG index | rag-corpus-builder |
| scrape thousands of known URLs | batch-url-scraping-jobs |
| map a site before crawling it | crawl-and-map-api-examples |
| pull property listings | zillow-listings-api-examples |
| compare hotel rates by market | hotels-and-travel-data-api |
Use-case pages: company data · price monitoring · market research · real estate · job postings
| Language | Where |
|---|---|
| Python | most repos — start with quanticdata-api-quickstart |
| Node.js | amazon-product-api-examples · youtube-scraper-api-examples · quanticdata-mcp-server |
| Go, PHP, Ruby, Java, C#, shell | quanticdata-api-quickstart · quanticdata-proxy-quickstart |
Pay per success. A call that fails costs nothing. Async jobs are charged on requested volume and auto-refund the unfetched share when they settle. Collectors bill per delivered row, so a run that returns nothing returns a bill of nothing.
Unit prices and the free monthly allowance: quanticdata.io/docs
curl converter · robots.txt tester · robots.txt generator · user-agent lookup
What is a web scraper API? · How a SERP API works · How MCP servers work · What is web data? · Is web scraping legal in the US? · …the whole blog
Every example repository is MIT licensed. Questions: hello@quanticdata.io · quanticdata.io · Partners

{ "type": "response", "message": "Scrape complete", "payload": { "url": "https://example.com", "markdown": "# Example Domain\n…", "metadata": { "title": "Example Domain", "canonical": "https://example.com/" }, "usage": { "cost_usd": 0.0002 } } }