Integration: Scavio

Search the web using Scavio, a unified search API for AI agents

Authors

Scavio

Table of Contents

• Overview
• Installation
• Usage
  • ScavioWebSearch
  • Parameters
• Beyond web search
• License

Overview

Scavio is a unified search API built for AI agents. One key reaches real-time data across Google, Amazon, Walmart, Target, eBay, Home Depot, YouTube, Reddit, TikTok, TikTok Shop, Instagram, X, Threads, Kuaishou, LinkedIn, Indeed, Glassdoor, Zillow, Redfin, Booking.com, Airbnb, Tripadvisor, Yelp, the Apple App Store, Google Play, SEC EDGAR, Companies House, G2, Capterra, Google Ads Transparency and the Meta Ad Library, plus an extract endpoint that turns any URL into HTML, Markdown or plain text. Everything comes back as structured JSON, with no scraping or proxies to run.

This integration is deliberately small. It ships one component:

• ScavioWebSearch: searches the web using Scavio’s Google web search endpoint and returns results as Haystack Document objects along with source URLs.

That makes it a drop-in peer of Haystack’s other web-search components, so it slots into an existing pipeline without rewiring. The rest of Scavio’s coverage is reachable from Haystack as well, but through the MCP server or the Python SDK rather than as one component per endpoint – see Beyond web search.

You need a Scavio API key to use this integration. You can get one at dashboard.scavio.dev.

Installation

pip install scavio-haystack

Usage

ScavioWebSearch

ScavioWebSearch queries the Scavio search API and returns results as Haystack Document objects containing the content snippets and metadata (title, URL). Source URLs are also returned separately.

Set your API key as the SCAVIO_API_KEY environment variable.

Basic Example

from haystack_integrations.components.websearch.scavio import ScavioWebSearch

web_search = ScavioWebSearch(top_k=5)

result = web_search.run(query="What is Haystack by deepset?")
documents = result["documents"]
links = result["links"]

Async Support

The component supports asynchronous execution via run_async:

import asyncio
from haystack_integrations.components.websearch.scavio import ScavioWebSearch

async def main():
    web_search = ScavioWebSearch(top_k=3)
    result = await web_search.run_async(query="What is Haystack by deepset?")
    print(f"Found {len(result['documents'])} documents")

asyncio.run(main())

Parameters

The component takes three arguments:

• api_key: API key for Scavio. Defaults to the SCAVIO_API_KEY environment variable.
• top_k: Maximum number of results to return. Defaults to 10.
• search_params: Additional parameters forwarded to the Scavio Google search endpoint. Set them once on the constructor, or pass search_params to run / run_async to replace them for a single call.

search_params keys

web_search = ScavioWebSearch(
    top_k=5,
    search_params={"gl": "de", "hl": "de", "time_period": "last_week"},
)

Legacy keys

Earlier versions of this page documented a different set of names. Three of them are still accepted and translated for you, so existing pipelines keep working:

Two more, search_type and light_request, belonged to a retired version of the endpoint. They are accepted but ignored – passing them has no effect on the request. Drop them.

Beyond web search

scavio-haystack wraps a single endpoint on purpose. To reach the rest of Scavio’s coverage from a Haystack agent, use either:

• the MCP server at mcp.scavio.dev, which exposes the full endpoint set as tools to any MCP-capable agent; or
• the Python SDK (pip install scavio), called from a custom Haystack component or a @tool-decorated function.

The endpoint reference lives at scavio.dev/docs.

License

scavio-haystack is distributed under the terms of the Apache-2.0 license.