GitHub Action

ScrapingAnt GitHub Action lets you scrape any URL directly from your GitHub Actions workflows — get HTML, Markdown, or AI-extracted structured data with rotating proxies and headless Chrome.

Installation

No installation required. Add the action to any workflow file in your repository:

- name: Scrape a webpage
  uses: scrapingant/scrape-action@v1
  with:
    api-key: ${{ secrets.SCRAPINGANT_API_KEY }}
    url: 'https://example.com'

Credentials setup

1. Register at ScrapingAnt and get your API key (10,000 free credits/month)
2. In your GitHub repository, go to Settings > Secrets and variables > Actions
3. Click New repository secret
4. Name it SCRAPINGANT_API_KEY and paste your API key

Available operations

The action supports three output modes via the output-type input:

Operation	API Endpoint	Description
HTML (html)	/v2/general	Scrape a webpage and return raw HTML. Supports headless Chrome, datacenter/residential proxies, and country targeting.
Markdown (markdown)	/v2/markdown	Convert a webpage to clean Markdown — optimized for LLM and RAG pipelines.
AI Extract (extract)	/v2/extract	Extract structured data from any page using natural language. Describe what you need and get back JSON.

Usage examples

HTML scraping

- name: Scrape HTML
  id: scrape
  uses: scrapingant/scrape-action@v1
  with:
    api-key: ${{ secrets.SCRAPINGANT_API_KEY }}
    url: 'https://example.com'
    output-type: 'html'
    browser: 'false'

- name: Use result
  run: echo "${{ steps.scrape.outputs.content }}" | head -20

Markdown for LLM/AI

- name: Get page as Markdown
  id: markdown
  uses: scrapingant/scrape-action@v1
  with:
    api-key: ${{ secrets.SCRAPINGANT_API_KEY }}
    url: 'https://example.com'
    output-type: 'markdown'

- name: Feed to AI
  run: echo "${{ steps.markdown.outputs.content }}"

AI data extraction

- name: Extract product data
  id: extract
  uses: scrapingant/scrape-action@v1
  with:
    api-key: ${{ secrets.SCRAPINGANT_API_KEY }}
    url: 'https://www.amazon.com/dp/B0EXAMPLE'
    output-type: 'extract'
    extract-properties: 'product title, price, rating, availability'

- name: Use extracted JSON
  run: echo "${{ steps.extract.outputs.content }}"

Save output to file

- name: Scrape and save
  uses: scrapingant/scrape-action@v1
  with:
    api-key: ${{ secrets.SCRAPINGANT_API_KEY }}
    url: 'https://example.com'
    output-type: 'markdown'
    output-file: 'scraped-content.md'

Inputs

Input	Required	Default	Description
api-key	Yes	—	ScrapingAnt API key
url	Yes	—	URL to scrape
output-type	No	html	Response type: html, markdown, or extract
extract-properties	No	—	Comma-separated fields to extract (for extract mode)
browser	No	true	Enable headless Chrome JS rendering
proxy-type	No	datacenter	Proxy type: datacenter or residential
proxy-country	No	—	Two-letter country code for geo-targeting (e.g. us, uk, de)
timeout	No	60	Max request time in seconds (5–60)
output-file	No	—	File path to save output

Outputs

Output	Description
content	Scraped content (HTML, Markdown, or JSON depending on output-type)
status-code	HTTP status code from ScrapingAnt API
url	Final URL after redirects (for markdown output-type)

Real-world use cases

• Competitor price monitoring — schedule a cron workflow to scrape product pages daily and save results to your repo
• Content change detection — scrape a page, compare with the previous version, and alert on diff
• AI-powered data pipeline — scrape, extract structured data, and feed to an LLM in the next workflow step
• SEO monitoring — scrape your pages as Markdown and check for content issues
• Post-deploy verification — verify your site renders correctly after each deployment

Credit costs

Credit costs vary by rendering mode and proxy type. See the credit cost reference for current pricing.

browser: true (default) costs 10 credits per request. browser: false costs 1 credit per request. Only successful responses are charged.

Resources

• GitHub Marketplace listing
• GitHub repository
• ScrapingAnt API documentation

Issues and tracking

In order to help us improve our service, please don't hesitate to create an issue or feature request at GitHub.