> ## Documentation Index
> Fetch the complete documentation index at: https://praison.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Web Search

> Give agents access to web search

Enable agents to search the web for current information.

```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
graph LR
    subgraph "Web Search"
        Q[❓ Question] --> A[🤖 Agent]
        A --> W[🌐 Web Search]
        W --> R[📄 Results]
        R --> A
        A --> O[💬 Answer]
    end
    
    classDef input fill:#6366F1,stroke:#7C90A0,color:#fff
    classDef agent fill:#F59E0B,stroke:#7C90A0,color:#fff
    classDef output fill:#10B981,stroke:#7C90A0,color:#fff
    
    class Q input
    class A,W,R agent
    class O output
```

## Quick Start

<Steps>
  <Step title="Create Agent with Web Search Tool">
    ```rust theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
    use praisonai::{Agent, tool};

    // Define web search tool
    #[tool(description = "Search the web for current information")]
    async fn web_search(query: String) -> String {
        // Use your preferred search API
        search_api::search(&query).await
    }

    // Create agent with web search capability
    let agent = Agent::new()
        .name("Researcher")
        .instructions("Search the web when asked about current events or news")
        .tool(web_search)
        .build()?;

    let response = agent.chat("What are today's top tech news?").await?;
    // Agent uses web_search tool and responds with current info
    ```
  </Step>

  <Step title="Using Multiple Search Tools">
    ```rust theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
    use praisonai::{Agent, tool};

    #[tool(description = "Search news articles")]
    async fn news_search(query: String) -> String {
        news_api::search(&query).await
    }

    #[tool(description = "Search general web")]
    async fn general_search(query: String) -> String {
        google_api::search(&query).await
    }

    let agent = Agent::new()
        .name("Researcher")
        .instructions("Use news_search for news, general_search for other queries")
        .tool(news_search)
        .tool(general_search)
        .build()?;
    ```
  </Step>
</Steps>

***

## How It Works

```mermaid theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
sequenceDiagram
    participant User
    participant Agent
    participant WebSearch
    
    User->>Agent: "Latest AI news?"
    Agent->>WebSearch: Search query
    WebSearch-->>Agent: Search results
    Agent->>Agent: Synthesize answer
    Agent-->>User: "Here are the latest..."
```

***

## Configuration

| Option        | Type    | Default | Description       |
| ------------- | ------- | ------- | ----------------- |
| `web_search`  | `bool`  | `false` | Enable web search |
| `max_results` | `usize` | `5`     | Results to fetch  |

***

## Related

<CardGroup cols={2}>
  <Card title="Tools" icon="wrench" href="/docs/rust/tools">
    Custom tools
  </Card>

  <Card title="Knowledge" icon="book" href="/docs/rust/knowledge">
    Local knowledge
  </Card>
</CardGroup>