SearchAPI | Agno

SearchApiTools gives an agent access to SearchAPI.io. Google web search is enabled by default. Enable News and YouTube individually. The current Images parser needs an upstream fix, so leave it disabled.

Prerequisites

The example requires Agno, the openai and requests packages, an OpenAI API key, and an API key from SearchAPI.io.

uv pip install -U agno openai requests

export OPENAI_API_KEY=***
export SEARCHAPI_API_KEY=***

Example

The following agent will search Google for the latest AI developments.

from agno.agent import Agent
from agno.tools.searchapi import SearchApiTools

agent = Agent(tools=[SearchApiTools()])
agent.print_response("What are the latest developments in AI agents?", markdown=True)

Toolkit Params

Parameter	Type	Default	Description
`api_key`	`Optional[str]`	`None`	SearchAPI key. If not provided, uses `SEARCHAPI_API_KEY` environment variable.
`num_results`	`int`	`5`	Default number of results to return per search.
`timeout`	`int`	`30`	Request timeout in seconds.
`enable_search_google`	`bool`	`True`	Enable Google web search.
`enable_search_news`	`bool`	`False`	Enable Google News search.
`enable_search_images`	`bool`	`False`	Enable Google Images search.
`enable_search_youtube`	`bool`	`False`	Enable YouTube search.
`all`	`bool`	`False`	Enable all available search functions.

Toolkit Functions

Function	Returns
`search_google(query: str, num_results: Optional[int] = None, location: Optional[str] = None, language: Optional[str] = None) -> str`	JSON string with organic results, the knowledge graph, related questions, and search information.
`search_news(query: str, num_results: Optional[int] = None, language: Optional[str] = None, country: Optional[str] = None) -> str`	JSON string with news titles, links, sources, dates, snippets, and thumbnails.
`search_images(query: str, num_results: Optional[int] = None, safe_search: Optional[str] = None) -> str`	Currently returns an empty result for the provider's `images` response; see compatibility below.
`search_youtube(query: str, num_results: Optional[int] = None) -> str`	JSON string with video IDs, titles, links, channels, lengths, views, publish times, descriptions, and thumbnails.

Google Images compatibility

The current Google Images response uses images. This adapter reads image_results, so it drops nonempty image results. Keep enable_search_images=False and avoid all=True until the parser is updated. The default web-search example is unaffected.

Developer Resources

Tools