Overview
Web search tools enable agents to find and retrieve information from the internet. This includes general search, news search, and advanced web scraping capabilities.
from agno.tools.duckduckgo import DuckDuckGoTools
Enable general web search functionality.
Enable news-specific search functionality.
Basic Usage
from agno.agent import Agent
from agno.models.nebius import Nebius
from agno.tools.duckduckgo import DuckDuckGoTools
import os
agent = Agent(
name="web_agent",
role="search the web for information based on user input",
model=Nebius(
id="deepseek-ai/DeepSeek-R1-0528",
api_key=os.getenv("NEBIUS_API_KEY")
),
tools=[DuckDuckGoTools(search=True, news=True)],
instructions=[
"You are a professional web search AI agent",
"Search the web for accurate, up-to-date information",
"Provide exact information available on the web"
]
)
response = agent.run("What are the latest developments in AI?")
print(response.content)
WebSearchService (TypeScript)
For TypeScript projects using the hedge fund example architecture:
import { WebSearchService } from './services/utils/WebSearchService';
apiKey
string
default:"process.env.SERPER_API_KEY"
API key for the Serper search service.
searchUrl
string
default:"'https://google.serper.dev/search'"
Base URL for the search API endpoint.
Constructor
const searchService = new WebSearchService(
apiKey?: string,
searchUrl?: string
);
Methods
search()
Performs a web search and returns structured results.
public async search(query: string): Promise<SearchResponse>
Whether the search was successful.
Error message if the search failed.
TypeScript Types
interface SearchResponse {
results: WebSearchResult[];
success: boolean;
error?: string;
}
interface WebSearchResult {
title: string;
snippet: string;
url: string;
}
Example Usage
import { WebSearchService } from './services/utils/WebSearchService';
const searchService = new WebSearchService(
process.env.SERPER_API_KEY,
'https://google.serper.dev/search'
);
const results = await searchService.search('AI news 2026');
if (results.success) {
results.results.forEach(result => {
console.log(`Title: ${result.title}`);
console.log(`Snippet: ${result.snippet}`);
console.log(`URL: ${result.url}`);
});
} else {
console.error(`Search failed: ${results.error}`);
}
from agno.tools.scrapegraph import ScrapeGraphTools
import os
scraper = ScrapeGraphTools(api_key=os.getenv("SGAI_API_KEY"))
ScrapeGraph AI API key for advanced web scraping.
Usage in Workflows
from agno.agent import Agent
from agno.models.nebius import Nebius
from agno.tools.scrapegraph import ScrapeGraphTools
from agno.workflow import Workflow
import os
class ResearchWorkflow(Workflow):
searcher: Agent = Agent(
tools=[ScrapeGraphTools(api_key=os.getenv("SGAI_API_KEY"))],
model=Nebius(
id="deepseek-ai/DeepSeek-V3-0324",
api_key=os.getenv("NEBIUS_API_KEY")
),
show_tool_calls=True,
markdown=True,
description="Expert at finding and extracting information from the web",
instructions=[
"Search for the most recent and authoritative sources",
"Extract key facts, statistics, and expert opinions",
"Cover multiple perspectives and highlight controversies",
"Include relevant statistics and data",
"Organize findings in a clear, structured format",
"Mention references and sources of the content"
]
)
from agno.agent import Agent
from agno.models.nebius import Nebius
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.tools.yfinance import YFinanceTools
import os
web_search_agent = Agent(
name="web_agent",
role="search the web for information",
model=Nebius(
id="deepseek-ai/DeepSeek-R1-0528",
api_key=os.getenv("NEBIUS_API_KEY")
),
tools=[DuckDuckGoTools(search=True, news=True)],
instructions=[
"You are a professional web search AI agent",
"Search the web for accurate information",
"Provide exact information available on the web"
]
)
financial_agent = Agent(
name="financial_agent",
role="get financial information",
model=Nebius(
id="Qwen/Qwen3-32B",
api_key=os.getenv("NEBIUS_API_KEY")
),
tools=[
YFinanceTools(
stock_price=True,
analyst_recommendations=True,
stock_fundamentals=True,
company_info=True,
technical_indicators=True,
historical_prices=True,
key_financial_ratios=True,
income_statements=True
)
],
instructions=[
"You are a professional financial advisor AI agent",
"Provide accurate financial information to users",
"Include stock prices, recommendations, and fundamentals"
]
)
multi_agent = Agent(
team=[web_search_agent, financial_agent],
model=Nebius(
id="meta-llama/Llama-3.3-70B-Instruct",
api_key=os.getenv("NEBIUS_API_KEY")
),
markdown=True
)
Search Configuration Options
Serper API Configuration
When using WebSearchService with Serper:
const response = await axios.post(searchUrl, {
q: query, // Search query
gl: 'us', // Geographic location
hl: 'en', // Language
num: 10 // Number of results
}, {
headers: {
'X-API-KEY': apiKey,
'Content-Type': 'application/json'
}
});
Result Types
The search service aggregates multiple result types:
- Organic results: Standard web search results
- People Also Ask: Related questions and answers
- Knowledge Graph: Structured information panels
Best Practices
- Enable appropriate search types: Only enable
search or news based on your needs
- Set clear instructions: Guide the agent on how to use search results
- Use markdown formatting: Enable
markdown=True for better result presentation
- Handle errors gracefully: Check
success field before processing results
- Respect rate limits: Implement delays for high-volume searches
- Cache results: Store frequently accessed search results to reduce API calls
Environment Variables
# For Serper API (WebSearchService)
SERPER_API_KEY=your_serper_api_key
# For ScrapeGraph AI (advanced scraping)
SGAI_API_KEY=your_scrapegraph_api_key
Error Handling
Python (Agno)
try:
response = agent.run("Search query")
print(response.content)
except Exception as e:
print(f"Search failed: {e}")
TypeScript
const result = await searchService.search(query);
if (!result.success) {
console.error(`Search error: ${result.error}`);
return;
}
if (result.results.length === 0) {
console.log('No results found');
return;
}
// Process results
result.results.forEach(item => {
console.log(item.title, item.url);
});
