> ## Documentation Index
> Fetch the complete documentation index at: https://resources.athenaintel.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Web Toolkit
Search the web and extract content with Athena's powerful Web Toolkit!
The Athena Web Toolkit provides a suite of tools for searching the internet and extracting content from web pages. Unlike the Browser toolkit which provides full Chrome automation, the Web Toolkit focuses on efficient web search and content retrieval without the overhead of managing browser sessions.
The Web Toolkit includes two primary tools:
* **Search**: Find information across the internet using intelligent web search
* **Browse**: Fetch and extract content from specific web pages and follow links
These tools are ideal for research tasks, content aggregation, competitive intelligence, and any workflow that requires gathering information from the web without complex browser interactions.
## Key Features
### Intelligent Web Search
Search the internet using natural language queries and receive relevant results with summaries. The Search tool understands context and can refine searches based on your requirements, making it easy to find specific information across the web.
### Content Extraction
Extract text, data, and structured information from web pages without managing a full browser session. The Browse tool fetches page content and intelligently extracts relevant information based on your instructions.
### Link Following
Automatically follow links from search results or web pages to gather information across multiple sources. The Browse tool can navigate through related pages to compile comprehensive information.
### Lightweight and Fast
Unlike full browser automation, the Web Toolkit operates without the overhead of rendering JavaScript or managing browser state, making it faster and more efficient for simple content retrieval tasks.
### Integration with Research Workflows
The Web Toolkit integrates seamlessly with other Athena capabilities including Deep Research, Notebooks, and Docs, enabling you to incorporate web content into your analysis and documentation workflows.
## Using the Search Tool
The Search tool enables you to find information across the internet using natural language queries. It's ideal for research, fact-checking, and gathering information on specific topics.
### When to Use Search
Use the Search tool when you need to:
* Find recent information or news about a topic
* Research companies, products, or technologies
* Gather multiple perspectives on a subject
* Locate specific types of content (articles, reports, documentation)
* Verify facts or find authoritative sources
### How Search Works
Describe what you're looking for in natural language. Be specific about the type of information you need and any constraints (date ranges, source types, etc.).
@Search for recent articles about artificial intelligence regulation in the European Union from the past 6 months
The Search tool returns relevant results with titles, URLs, and summaries. Review these to identify the most promising sources for your needs.
If initial results don't meet your needs, refine your query with additional context, constraints, or alternative phrasing. The Search tool learns from your refinements to improve results.
Once you've identified relevant sources, use the Browse tool to extract detailed content from specific pages for deeper analysis.
## Using the Browse Tool
The Browse tool fetches content from specific web pages and extracts relevant information. Unlike the Browser toolkit which provides full Chrome automation, Browse focuses on efficient content retrieval.
### When to Use Browse
Use the Browse tool when you need to:
* Extract text content from specific web pages
* Gather information from multiple pages on a website
* Follow links to compile information across related pages
* Scrape structured data from simple web pages
* Read articles, documentation, or reports from known URLs
### How Browse Works
Provide the URL of the page you want to access, or ask Browse to follow links from search results or other pages.
@Browse [https://www.reuters.com/](https://www.reuters.com/) and extract the main content, author, and publication date of the latest article
Clearly specify what information you want to extract from the page. Browse will focus on the relevant content and ignore navigation, ads, and other peripheral elements.
If you need information from multiple pages, Browse can follow links and aggregate content. Specify the navigation pattern or link criteria to follow.
Browse presents extracted content in a structured format suitable for your needs—whether that's a summary, a detailed extraction, or structured data ready for analysis.
## Choosing the Right Tool
Understanding when to use Search, Browse, or the Browser toolkit helps you select the most efficient approach for your web-based tasks.
### Use Search When:
* You need to find information but don't know specific sources
* You're researching a topic and need multiple perspectives
* You want recent news or updates on a subject
* You need to identify authoritative sources on a topic
* You're starting a research project and need to discover relevant content
### Use Browse When:
* You have specific URLs to extract content from
* You need to read articles or documentation from known sources
* You want to extract structured data from simple web pages
* You need to follow links within a website to gather related information
* The target pages are publicly accessible and don't require authentication
### Use Browser Toolkit When:
* You need to interact with forms, buttons, or other UI elements
* The target website requires authentication or login
* Content is dynamically loaded via JavaScript
* You need to maintain session state across multiple visits
* You need to automate complex multi-step workflows on web applications
* You require session recording for compliance or audit purposes
### Combining Tools
For many research workflows, you'll use multiple tools together:
1. **Search** to discover relevant sources
2. **Browse** to extract content from simple, public pages
3. **Browser** for authenticated access or complex interactions
4. **Deep Research** to orchestrate comprehensive research across web and internal sources