SerperTools enable an Agent to search Google, search news, search academic papers, search reviews, and scrape webpages using Serper.

​
Prerequisites

The following example requires an API key from Serper. 
export SERPER_API_KEY=***

​
Example

The following agent will search for the latest news about artificial intelligence developments:
cookbook/tools/serper_tools.py
from agno.agent import Agent
from agno.tools.serper import SerperTools

agent = Agent(
    tools=[SerperTools()],
    show_tool_calls=True,
)

agent.print_response(
    "Search for the latest news about artificial intelligence developments",
    markdown=True,
)

​
Additional Examples

agent.print_response(
    "Find 2 recent academic papers about large language model safety and alignment",
    markdown=True,
)

agent.print_response(
    "Use this Google Place ID: ChIJ_Yjh6Za1j4AR8IgGUZGDDTs and analyze the reviews",
    markdown=True
)

​
Web Scraping

agent.print_response(
    "Scrape and summarize the main content from this OpenAI blog post: https://openai.com/index/gpt-4/",
    markdown=True
)

​
Toolkit Params

ParameterTypeDefaultDescription
api_keyOptional[str]NoneAPI key for authentication.
countrystr"us"Country code for search results (e.g., “us”, “uk”).
locationOptional[str]NoneGoogle location code for search results.
languagestr"en"Language code for search results (e.g., “en”, “es”).
num_resultsint10Default number of search results to retrieve.
date_rangeOptional[str]NoneDefault date range filter for searches.
sort_reviews_byOptional[str]"mostRelevant"Sort order for reviews search (“mostRelevant”, “newest”, etc.).

​
Toolkit Functions

FunctionDescription
searchSearches Google for a query. Parameters: query (str), optional num_results (int).
search_newsSearches for news articles. Parameters: query (str), optional num_results (int).
search_scholarSearches Google Scholar for academic papers. Parameters: query (str), optional num_results (int).
search_reviewsSearches for reviews using place identifiers. Parameters: optional place_id, cid, fid, or topic_id (str).
scrape_webpageScrapes content from a webpage. Parameters: url (str), optional markdown (bool).

​
Developer Resources