Residential, Mobile, ISP & Datacenter proxies, plus Scraping APIs
— built for AI pipelines, growth teams, and large-scale web data collection.

Website • Docs • Dashboard • Python SDK (PyPI)

Thordata provides a full‑stack web data platform:

All proxies are exposed via a simple HTTP/HTTPS gateway.

Official SDKs and low‑level clients for accessing Thordata products.

• thordata-python-sdk
  Modern Python SDK (published as thordata-sdk) with sync & async clients for: Residential / Datacenter / Mobile proxies, SERP API, Universal Scraper API, and Web Scraper task management.

Planned:

• thordata-node-sdk — Node.js SDK for proxies + SERP
• thordata-go-sdk — Go SDK for proxy & scraping workloads

Tools and examples that connect Thordata with AI agents, RAG pipelines, and model tool ecosystems.

• thordata-cookbook
  A collection of end‑to‑end recipes:

  • RAG data pipeline with Universal Scraper → HTML cleaning → Markdown
  • Web QA Agent: question → SERP search → page scraping → LLM answer
  • MCP tools: expose search_web, search_news, read_website, extract_links to LLMs
  • GitHub repository intelligence and app‑store review analysis

• thordata-langchain-tools
  LangChain tools powered by Thordata:

  • ThordataSerpTool — real‑time web search via SERP API
  • ThordataScrapeTool — universal single‑page scraping with optional JS rendering

• thordata-web-qa-agent
  CLI Web Q&A agent: question → Thordata SERP → Universal Scraper → HTML cleaning → OpenAI answer.

Quick‑start examples for using Thordata's proxy network.

• thordata-proxy-examples
  Minimal examples showing how to:
  • Send HTTP requests via Thordata Residential / Mobile / Datacenter proxies (Python + curl)
  • Configure basic geo‑targeting (country / city‑level)
  • Run concurrent IP checks and simple health monitoring

Planned:

• thordata-proxy-docker — dockerized local forward proxy using Thordata credentials

SERP‑based examples focused on Google and news use cases.

• google-news-scraper
  Full‑featured CLI example for engine=google_news, supporting:

  • q (query), hl (language), gl (country)
  • topic_token, publication_token, section_token, story_token, so
  • CSV export of structured news results via Thordata SERP API

• google-play-reviews-rag
  Google Play app reviews analysis + RAG: fetch reviews via Thordata Web Scraper, build an embeddings index, and answer questions about user sentiment.

Planned:

• Generic Google web search examples (engine=google)
• Google Maps / Play / Shopping examples as separate repositories

Hands‑on guides and notebooks to help you build data pipelines on top of Thordata.

Most of these live in thordata-cookbook:

• notebooks/rag/rag_openai_research.ipynb —
  Prepare dynamic HTML content (e.g. OpenAI Research) for RAG by scraping, cleaning, and exporting to Markdown.

• notebooks/devtools/github_repo_intel.ipynb —
  Use Web Scraper API spiders to collect GitHub repository metadata (stars, issues, contributors, languages) into a Pandas DataFrame.

• notebooks/ai/web_qa_agent_with_thordata.ipynb —
  End‑to‑end "Web Q&A Agent": question → SERP search → Universal Scraper → HTML cleaning → LLM answer.

All notebooks support:

• Live mode — call Thordata APIs and cache results under data/
• Offline mode — reuse cached HTML/JSON without consuming credits

pip install thordata-sdk

from thordata import ThordataClient

client = ThordataClient(
    scraper_token="YOUR_SCRAPER_TOKEN",
    public_token="YOUR_PUBLIC_TOKEN",
    public_key="YOUR_PUBLIC_KEY",
)

resp = client.get("http://httpbin.org/ip")
print(resp.json())  # → see your Thordata exit IP

from thordata import Engine

results = client.serp_search(
    query="Thordata proxy network",
    engine=Engine.GOOGLE,
    num=5,
    # Pass engine‑specific params via **kwargs, e.g.:
    # engine="google_news", location="United States"
)

print("Organic results:", len(results.get("organic", [])))

html = client.universal_scrape(
    url="https://www.thordata.com",
    js_render=True,
    output_format="html",
)
print(html[:500])

For more advanced examples (Web Scraper tasks, async high‑concurrency, RAG pipelines, MCP tools), see:

• SDK examples → thordata-python-sdk/examples
• Cookbook scripts & notebooks → thordata-cookbook

• Dashboard: https://www.thordata.com/
• Docs: https://doc.thordata.com
• Python SDK: https://github.com/Thordata/thordata-python-sdk
• Cookbook: https://github.com/Thordata/thordata-cookbook
• Support: [email protected]

If you are building something interesting on top of Thordata (RAG pipelines, AI agents, dashboards), feel free to open an issue and share your project — we are happy to feature selected community examples.

Thordata powers the proxy network and web data pipelines behind modern AI.

_{Last updated: 2025‑12‑01}