Skip to main content

​
Quickstart

DumplingAI is a data extraction and automation API. Use it to pull structured data out of websites, PDFs, YouTube videos, TikTok, LinkedIn, Google Search, and more, without managing scrapers, proxies, or third-party integrations. You can use the existing endpoint-oriented API or the newer agent-oriented /api/v2 platform for capability-first execution and provider endpoints. This page combines the product overview with the fastest path to your first successful request.

​
Choose your path

MCP Server

Connect DumplingAI to Claude, Cursor, Windsurf, VS Code, or n8n

CLI & Skills

Use DumplingAI from the terminal and install coding-agent skill packs

API Reference

Browse endpoints, parameters, sample requests, and credit costs

​
1. Create an API key

  1. Sign up for a DumplingAI account
  2. Navigate to the API Keys section in your dashboard
  3. Click “Create API Key” and give it a descriptive name
  4. Copy your API key for use in API requests and MCP connections
Keep your API key secure. Do not share it or expose it in client-side code.

​
2. Send a test request

Start with the Agent API v2 search_news capability to verify your setup: 
curl -X POST https://app.dumplingai.com/api/v2/run \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "type": "capability",
    "id": "search_news",
    "input": {
      "query": "latest AI infrastructure news",
      "dateRange": "pastWeek"
    }
  }'

​
3. Understand credits

DumplingAI uses credits to track API usage.
  • Search capabilities like google_search and search_news start at 10-30 credits
  • Higher-value extraction capabilities like extract_web and extract_document cost more because they include AI parsing and document processing
See Credit Costs for the full breakdown, and Agent API v2 for the capability-first surface.

​
What you can build

Web Scraping & Crawling

Scrape pages, crawl entire sites, take screenshots, and extract structured data using AI

Social Media Data

Fetch transcripts, profiles, videos, and comments from YouTube, TikTok, and LinkedIn

Search

Web search, news, maps, places, Google autocomplete, and business reviews

Document Processing

Convert, merge, and extract data from PDFs, images, audio, and video files

AI Agents

Automate complex research and data tasks with AI agents

Code Execution

Run JavaScript and Python securely in a sandbox environment

​
Next steps

Web Scraping Tutorial

Learn how to scrape sites, clean content, and extract structured data

Agent API

Build multi-step workflows with the agent completion endpoint

MCP Server Setup

Connect DumplingAI to your AI client instead of calling the API directly

API Reference

Explore endpoint details, parameters, examples, and response formats