QuanticData Data API bronze

Scrape a URL

Fetch one page through the proxy network and return it as markdown, html or text, optionally with a structured extraction schema (CSS selectors), a screenshot or browser actions. Same parameters as POST /v1/scrape (url, format, engine, render, country, extract, actions, screenshot …). Completes in the same call.

• {"skill":"scrape_url","input":{"url":"https://example.com/pricing","format":"markdown","country":"us"}}
• {"skill":"scrape_url","input":{"url":"https://example.com/pricing","extract":{"title":"h1","price":".price"}}}

Web search (SERP)

Structured search results from Google, Bing or DuckDuckGo: organic results, shopping, images, news, places, maps, jobs, autocomplete and the other verticals. Same parameters as POST /v1/serp (query, engine, search_type, country, lang, num, page, device, location …). Completes in the same call.

• {"skill":"web_search","input":{"query":"best espresso machines 2026","country":"us","num":10}}
• {"skill":"web_search","input":{"query":"pizzeria napoli","search_type":"places","country":"it","lang":"it"}}

Map a site

Discover the URLs of a site from its sitemaps and homepage links, with per-section counts. Same parameters as POST /v1/map (url, limit, search, includeSubdomains, sitemapOnly, group_by, country). Completes in the same call.

• {"skill":"map_site","input":{"url":"https://example.com","limit":100,"search":"blog"}}

Crawl a site

Crawl up to 500 pages from a seed URL and return each page as markdown, html or text. Same parameters as POST /v1/crawl (url, limit, depth, format, contentMode, render, include, exclude, country). Asynchronous: the task stays TASK_STATE_WORKING while the job runs; poll it with GetTask, the crawled pages arrive as an artifact when it is TASK_STATE_COMPLETED. CancelTask stops the job.

• {"skill":"crawl_site","input":{"url":"https://docs.example.com","limit":50,"depth":3,"include":["/guides/*"]}}

Run a collector

Run one of the ready-made collectors (Amazon, Google Maps, app stores, job boards, …) with its semantic input and get typed rows back. Input: {"slug": "<collector slug>", ...collector input} — same body as POST /v1/scraper/collectors/{slug}/run; the catalog with each collector's input_schema is GET /v1/scraper/collectors. Short runs complete in the same call; long runs return a TASK_STATE_WORKING task to poll with GetTask. Collector runs cannot be canceled.

• {"skill":"run_collector","input":{"slug":"amazon_search","keyword":"espresso machine","country":"us","max_results":20}}
• {"skill":"run_collector","input":{"slug":"google_maps_places","query":"dentist","location":"Milan, Italy","async":true}}

Build a dataset

Build a dataset from a natural-language prompt: the pipeline searches, maps and scrapes the sources and returns validated rows with the requested columns. Same parameters as POST /v1/scraper/datasets (prompt, columns, country, sources, limits, webhook, refresh, refreshUrls). Asynchronous: poll the task with GetTask; the rows and download links arrive as an artifact at TASK_STATE_COMPLETED. CancelTask stops the job.

• {"skill":"build_dataset","input":{"prompt":"Italian B2B SaaS companies with pricing pages","columns":["name","website","pricing_url"],"country":"it","limits":{"max_rows":100}}}

QuanticData MCP server

QuanticData web data API (scrape, search, crawl, unlock, collectors, datasets) and proxies as 29 MCP tools: scrape, search, video reading, map, crawl, batch, unlock, AI visibility, SEO audit, collectors, datasets, parser presets, proxies. Same backend as the REST Data API (https://api.quanticdata.io/v1).

QuanticData Data API (OpenAPI 3.1)

REST API at https://api.quanticdata.io/v1 with 48 operations: scrape, search, map, crawl, batch, unlock, AI extraction, collectors, datasets, parser presets, proxies, account. Bearer API key.

QuanticData API reference for agents (llms.txt)

Index of the whole API reference in Markdown, one line per endpoint, with links to the per-section Markdown files.

Use data api

Call the QuanticData Data API (REST, https://api.quanticdata.io/v1): authenticate with an API key, read the response envelope, run async jobs, and find the OpenAPI and Markdown references. Use when an agent needs to scrape, search, crawl, unlock, run collectors, build datasets or manage proxies over plain HTTP.

Use mcp server

Connect an MCP client (Claude, Cursor, ChatGPT, any MCP host) to the QuanticData MCP server: remote Streamable HTTP endpoint with OAuth 2.1 or API key, local npm package quanticdata-mcp, tool presets, and the list of the 29 tools. Use when the user wants QuanticData tools inside an AI assistant instead of raw HTTP calls.

Scrape to markdown

Turn any web page into clean Markdown, HTML or text with QuanticData: choose the fetch engine, keep only the relevant section of a large page, extract structured fields with CSS selectors or an AI prompt, and scale to many URLs with batch, map and crawl. Use when an agent needs page content it cannot fetch itself (JavaScript, blocks, geo-restricted pages).