Riding the Web with WebSurferAgent

Author:

Robert Jambrecic

Machine Learning Engineer at Airt

Introduction

In our Adding Browsing Capabilities to AG2 guide, we explored how to build agents with basic web surfing capabilities. Now, let’s take it to the next level with WebSurferAgent—a powerful agent that comes with built-in web browsing tools right out of the box! With WebSurferAgent, your agents can seamlessly browse the web, retrieve real-time information, and interact with web pages—all with minimal setup. WebSurferAgent Example

WebSurferAgent with BrowserUseTool

Browser Use requires Python 3.11 or higher.

Installation

To get started with the Browser Use integration in AG2, follow these steps:

Install AG2 with the browser-use extra:
```
pip install -U ag2[openai,browser-use]
```
If you have been using autogen or ag2, all you need to do is upgrade it using:
pip install -U autogen[openai,browser-use]
or
pip install -U ag2[openai,browser-use]
as autogen and ag2 are aliases for the same PyPI package.

Set up Playwright:

# Installs Playwright and browsers for all OS
playwright install
# Additional command, mandatory for Linux only
playwright install-deps

For running the code in Jupyter, use nest_asyncio to allow nested event loops.
```
pip install nest_asyncio
```

You’re all set! Now you can start using browsing features in AG2.

Imports

import os

import nest_asyncio

from autogen import LLMConfig
from autogen.agentchat import UserProxyAgent
from autogen.agents.experimental import WebSurferAgent

nest_asyncio.apply()

Configure WebSurferAgent with Browser Use

Browser Use supports the following models: Supported ModelsWe had great experience with OpenAI, Anthropic, and Gemini. However, DeepSeek and Ollama haven’t performed as well.

llm_config = LLMConfig(api_type="openai", model="gpt-4o-mini", api_key=os.environ["OPENAI_API_KEY"])

WebSurferAgent is the one responsible for browsing the web and retrieving information. The web_tool="browser_use" tells the agent to use the BrowserUseTool to surf the web. After creating the WebSurferAgent there are two ways to start the chat session:

Recommended: Using the `run` Method

The new run method simplifies the process by eliminating the need for manual UserProxyAgent creation.

✅ Easier setup – No need to manually register tools

with llm_config:
    websurfer = WebSurferAgent(name="WebSurfer", web_tool="browser_use")

run_response = websurfer.run(
    message="Get info from https://docs.ag2.ai/docs/Home",
    tools=websurfer.tools,
    max_turns=2,
    user_input=False,
)

# Iterate through the chat automatically with console output
run_response.process()

Manual Setup: Using `initiate_chat` Method

This method requires manually creating a UserProxyAgent and registering tools for execution.

⚠️ More setup required
⚠️ Must manually register tools

with llm_config:
    websurfer = WebSurferAgent(name="WebSurfer", web_tool="browser_use")
user_proxy = UserProxyAgent(name="user_proxy", human_input_mode="NEVER")
# WebSurferAgent has a list of tools which are registered for LLM
# We need to register the tools for execution with the UserProxyAgent
for tool in websurfer.tools:
    tool.register_for_execution(user_proxy)

user_proxy.initiate_chat(
    recipient=websurfer,
    message="Get info from https://docs.ag2.ai/docs/Home",
    max_turns=2,
)

Output

user_proxy (to WebSurfer):

Get info from https://docs.ag2.ai/docs/Home

--------------------------------------------------------------------------------

>>>>>>>> USING AUTO REPLY...
WebSurfer (to user_proxy):

***** Suggested tool call (call_rl777jGrOGhc68goW5142urK): browser_use *****
Arguments:
{"task":"Get info from https://docs.ag2.ai/docs/Home"}
****************************************************************************

--------------------------------------------------------------------------------

>>>>>>>> EXECUTING FUNCTION browser_use...
Call ID: call_rl777jGrOGhc68goW5142urK
Input arguments: {'task': 'Get info from https://docs.ag2.ai/docs/Home'}
INFO     [agent] 🚀 Starting task: Get info from https://docs.ag2.ai/docs/Home
INFO     [agent]
📍 Step 1
INFO     [agent] 🤷 Eval: Unknown - No previous actions to evaluate.
INFO     [agent] 🧠 Memory:
INFO     [agent] 🎯 Next goal: Navigate to the specified URL to gather information
INFO     [agent] 🛠️  Action 1/1: {"go_to_url":{"url":"https://docs.ag2.ai/docs/Home"}}
INFO     [controller] 🔗  Navigated to https://docs.ag2.ai/docs/Home
INFO     [agent]
📍 Step 2
INFO     [agent] 🤷 Eval: Unknown - No previous actions to evaluate.
INFO     [agent] 🧠 Memory:
INFO     [agent] 🎯 Next goal: Extract useful information regarding AG2 from the home page.
INFO     [agent] 🛠️  Action 1/1: {"extract_content":{"include_links":true}}
INFO     [controller] 📄  Extracted page as markdown
: [AG2 home page![light logo](https://mintlify.s3.us-
west-1.amazonaws.com/ag2ai/logo/ag2.svg)![dark logo](https://mintlify.s3.us-
west-1.amazonaws.com/ag2ai/logo/ag2-white.svg)](/)

Search or ask...

⌘K

Search...A

Navigation

Home

AG2

[Home](/docs/home/Home)[User Guide](/docs/user-guide/quick-start)[API
References](/docs/api-reference/autogen/overview)[Use Cases](/docs/use-
cases/use-cases/customer-service)[Contribute](/contributor-
guide/contributing)[FAQs](/faq/FAQ)[Ecosystem](/ecosystem/agentops)[Blog](/docs/blog/2025-01-29-RealtimeAgent-
with-gemini/index)

##### Home

  * [AG2](/docs/home/Home)

![AG2 Logo](https://mintlify.s3.us-
west-1.amazonaws.com/ag2ai/static/img/ag2.svg)

## AG2

The Open Source Agent OS

[Getting Started - 3 Minute](/docs/user-guide/quick-start)

###



Key Features

##

![Multi-Agent Conversation Framework](https://mintlify.s3.us-
west-1.amazonaws.com/ag2ai/static/img/conv_2.svg)**Multi-Agent Conversation
Framework**

AG2 provides multi-agent conversation framework as a high-level abstraction.
With this framework, one can conveniently build LLM workflows.

##

![Easily Build Diverse Applications](https://mintlify.s3.us-
west-1.amazonaws.com/ag2ai/static/img/autogen_app.svg)**Easily Build Diverse
Applications**

AG2 offers a collection of working systems spanning a wide range of
applications from various domains and complexities.

##

![Enhanced LLM Inference & Optimization](https://mintlify.s3.us-
west-1.amazonaws.com/ag2ai/static/img/extend.svg)**Enhanced LLM Inference &
Optimization**

AG2 supports enhanced LLM inference APIs, which can be used to improve
inference performance and reduce cost.

###



Explore content

## [Quick StartStart building your AG2 application.](/docs/user-guide/quick-
start)## [ConceptsWork through the key concepts of AG2 including
ConversableAgent, GroupChat, Swarm, and tools.](/docs/user-guide/basic-
concepts/installing-ag2)## [Advanced ConceptsAdvance to RAG, Code Execution,
and more complex GroupChats and Swarms.](/docs/user-guide/advanced-concepts)##
[Use CasesTry out use case workflows including Customer Service, Travel
Planning, and Game Design.](/docs/use-cases/use-cases)## [Notebook ExamplesA
collection of interactive notebooks across all AG2 topics.](/docs/use-
cases/notebooks/Notebooks)## [API ReferenceDelve into the AG2 API
reference.](/docs/api-reference)## [How to ContributeGet involved with AG2 by
adding what you need and making the framework even better!](/contributor-
guide/contributing)

###



Popular resources

##

[Foundation Capital Interview with Dr. Chi
Wang](https://www.youtube.com/watch?v=RLwyXRVvlNk)

##

[Learn AG2 on DeepLearningAI](https://www.youtube.com/watch?v=TBNTH-fwGPE)

On this page

  * Key Features
  * Explore content
  * Popular resources



INFO     [agent]
📍 Step 3
INFO     [agent] 👍 Eval: Success - Successfully extracted information from the AG2 home page.
INFO     [agent] 🧠 Memory: Extracted information includes the key features and content related to AG2.
INFO     [agent] 🎯 Next goal: Complete the task and summarize the extracted information.
INFO     [agent] 🛠️  Action 1/1: {"done":{"text":"Extracted information from AG2 home page:\n\n### AG2\n\n**The Open Source Agent OS**  \n[Getting Started - 3 Minute](https://docs/user-guide/quick-start)\n\n#### Key Features\n\n- **Multi-Agent Conversation Framework**: AG2 provides a high-level abstraction for building LLM workflows.\n- **Easily Build Diverse Applications**: Offers a collection of systems across various domains and complexities.\n- **Enhanced LLM Inference & Optimization**: Supports APIs to improve inference performance and reduce costs.\n\n#### Explore Content\n- [Quick Start](https://docs/user-guide/quick-start)\n- [Concepts](https://docs/user-guide/basic-concepts/installing-ag2)\n- [Advanced Concepts](https://docs/user-guide/advanced-concepts)\n- [Use Cases](https://docs/use-cases/use-cases)\n- [API Reference](https://docs/api-reference)\n- [How to Contribute](https://contributor-guide/contributing)  \n\n#### Popular Resources\n- [Foundation Capital Interview with Dr. Chi Wang](https://www.youtube.com/watch?v=RLwyXRVvlNk)  \n- [Learn AG2 on DeepLearningAI](https://www.youtube.com/watch?v=TBNTH-fwGPE)  \n"}}
INFO     [agent] 📄 Result: Extracted information from AG2 home page:

### AG2

**The Open Source Agent OS**
[Getting Started - 3 Minute](https://docs/user-guide/quick-start)

#### Key Features

- **Multi-Agent Conversation Framework**: AG2 provides a high-level abstraction for building LLM workflows.
- **Easily Build Diverse Applications**: Offers a collection of systems across various domains and complexities.
- **Enhanced LLM Inference & Optimization**: Supports APIs to improve inference performance and reduce costs.

#### Explore Content
- [Quick Start](https://docs/user-guide/quick-start)
- [Concepts](https://docs/user-guide/basic-concepts/installing-ag2)
- [Advanced Concepts](https://docs/user-guide/advanced-concepts)
- [Use Cases](https://docs/use-cases/use-cases)
- [API Reference](https://docs/api-reference)
- [How to Contribute](https://contributor-guide/contributing)

#### Popular Resources
- [Foundation Capital Interview with Dr. Chi Wang](https://www.youtube.com/watch?v=RLwyXRVvlNk)
- [Learn AG2 on DeepLearningAI](https://www.youtube.com/watch?v=TBNTH-fwGPE)

INFO     [agent] ✅ Task completed successfully
user_proxy (to WebSurfer):

***** Response from calling tool (call_rl777jGrOGhc68goW5142urK) *****
{"extracted_content":["🔗  Navigated to https://docs.ag2.ai/docs/Home","📄  Extracted page as markdown\n: [AG2 home page![light logo](https://mintlify.s3.us-\nwest-1.amazonaws.com/ag2ai/logo/ag2.svg)![dark logo](https://mintlify.s3.us-\nwest-1.amazonaws.com/ag2ai/logo/ag2-white.svg)](/)\n\nSearch or ask...\n\n⌘K\n\nSearch...\n\nNavigation\n\nHome\n\nAG2\n\n[Home](/docs/home/Home)[User Guide](/docs/user-guide/quick-start)[API\nReferences](/docs/api-reference/autogen/overview)[Use Cases](/docs/use-\ncases/use-cases/customer-service)[Contribute](/contributor-\nguide/contributing)[FAQs](/faq/FAQ)[Ecosystem](/ecosystem/agentops)[Blog](/docs/blog/2025-01-29-RealtimeAgent-\nwith-gemini/index)\n\n##### Home\n\n  * [AG2](/docs/home/Home)\n\n![AG2 Logo](https://mintlify.s3.us-\nwest-1.amazonaws.com/ag2ai/static/img/ag2.svg)\n\n## AG2\n\nThe Open Source Agent OS\n\n[Getting Started - 3 Minute](/docs/user-guide/quick-start)\n\n###\n\n\n\nKey Features\n\n##\n\n![Multi-Agent Conversation Framework](https://mintlify.s3.us-\nwest-1.amazonaws.com/ag2ai/static/img/conv_2.svg)**Multi-Agent Conversation\nFramework**\n\nAG2 provides multi-agent conversation framework as a high-level abstraction.\nWith this framework, one can conveniently build LLM workflows.\n\n##\n\n![Easily Build Diverse Applications](https://mintlify.s3.us-\nwest-1.amazonaws.com/ag2ai/static/img/autogen_app.svg)**Easily Build Diverse\nApplications**\n\nAG2 offers a collection of working systems spanning a wide range of\napplications from various domains and complexities.\n\n##\n\n![Enhanced LLM Inference & Optimization](https://mintlify.s3.us-\nwest-1.amazonaws.com/ag2ai/static/img/extend.svg)**Enhanced LLM Inference &\nOptimization**\n\nAG2 supports enhanced LLM inference APIs, which can be used to improve\ninference performance and reduce cost.\n\n###\n\n\n\nExplore content\n\n## [Quick StartStart building your AG2 application.](/docs/user-guide/quick-\nstart)## [ConceptsWork through the key concepts of AG2 including\nConversableAgent, GroupChat, Swarm, and tools.](/docs/user-guide/basic-\nconcepts/installing-ag2)## [Advanced ConceptsAdvance to RAG, Code Execution,\nand more complex GroupChats and Swarms.](/docs/user-guide/advanced-concepts)##\n[Use CasesTry out use case workflows including Customer Service, Travel\nPlanning, and Game Design.](/docs/use-cases/use-cases)## [Notebook ExamplesA\ncollection of interactive notebooks across all AG2 topics.](/docs/use-\ncases/notebooks/Notebooks)## [API ReferenceDelve into the AG2 API\nreference.](/docs/api-reference)## [How to ContributeGet involved with AG2 by\nadding what you need and making the framework even better!](/contributor-\nguide/contributing)\n\n###\n\n\n\nPopular resources\n\n##\n\n[Foundation Capital Interview with Dr. Chi\nWang](https://www.youtube.com/watch?v=RLwyXRVvlNk)\n\n##\n\n[Learn AG2 on DeepLearningAI](https://www.youtube.com/watch?v=TBNTH-fwGPE)\n\nOn this page\n\n  * Key Features\n  * Explore content\n  * Popular resources\n\n\n","Extracted information from AG2 home page:\n\n### AG2\n\n**The Open Source Agent OS**  \n[Getting Started - 3 Minute](https://docs/user-guide/quick-start)\n\n#### Key Features\n\n- **Multi-Agent Conversation Framework**: AG2 provides a high-level abstraction for building LLM workflows.\n- **Easily Build Diverse Applications**: Offers a collection of systems across various domains and complexities.\n- **Enhanced LLM Inference & Optimization**: Supports APIs to improve inference performance and reduce costs.\n\n#### Explore Content\n- [Quick Start](https://docs/user-guide/quick-start)\n- [Concepts](https://docs/user-guide/basic-concepts/installing-ag2)\n- [Advanced Concepts](https://docs/user-guide/advanced-concepts)\n- [Use Cases](https://docs/use-cases/use-cases)\n- [API Reference](https://docs/api-reference)\n- [How to Contribute](https://contributor-guide/contributing)  \n\n#### Popular Resources\n- [Foundation Capital Interview with Dr. Chi Wang](https://www.youtube.com/watch?v=RLwyXRVvlNk)  \n- [Learn AG2 on DeepLearningAI](https://www.youtube.com/watch?v=TBNTH-fwGPE)  \n"],"final_result":"Extracted information from AG2 home page:\n\n### AG2\n\n**The Open Source Agent OS**  \n[Getting Started - 3 Minute](https://docs/user-guide/quick-start)\n\n#### Key Features\n\n- **Multi-Agent Conversation Framework**: AG2 provides a high-level abstraction for building LLM workflows.\n- **Easily Build Diverse Applications**: Offers a collection of systems across various domains and complexities.\n- **Enhanced LLM Inference & Optimization**: Supports APIs to improve inference performance and reduce costs.\n\n#### Explore Content\n- [Quick Start](https://docs/user-guide/quick-start)\n- [Concepts](https://docs/user-guide/basic-concepts/installing-ag2)\n- [Advanced Concepts](https://docs/user-guide/advanced-concepts)\n- [Use Cases](https://docs/use-cases/use-cases)\n- [API Reference](https://docs/api-reference)\n- [How to Contribute](https://contributor-guide/contributing)  \n\n#### Popular Resources\n- [Foundation Capital Interview with Dr. Chi Wang](https://www.youtube.com/watch?v=RLwyXRVvlNk)  \n- [Learn AG2 on DeepLearningAI](https://www.youtube.com/watch?v=TBNTH-fwGPE)  \n"}
**********************************************************************

--------------------------------------------------------------------------------

>>>>>>>> USING AUTO REPLY...
WebSurfer (to user_proxy):

Here's the information extracted from the AG2 home page:

### AG2
**The Open Source Agent OS**
[Getting Started - 3 Minute](https://docs/user-guide/quick-start)

#### Key Features
- **Multi-Agent Conversation Framework**: AG2 provides a high-level abstraction for building LLM (Large Language Model) workflows.
- **Easily Build Diverse Applications**: Offers a collection of systems across various domains and complexities.
- **Enhanced LLM Inference & Optimization**: Supports enhanced inference APIs to improve performance and reduce costs.

#### Explore Content
- [Quick Start](https://docs/user-guide/quick-start)
- [Concepts](https://docs/user-guide/basic-concepts/installing-ag2)
- [Advanced Concepts](https://docs/user-guide/advanced-concepts)
- [Use Cases](https://docs/use-cases/use-cases)
- [API Reference](https://docs/api-reference)
- [How to Contribute](https://contributor-guide/contributing)

#### Popular Resources
- [Foundation Capital Interview with Dr. Chi Wang](https://www.youtube.com/watch?v=RLwyXRVvlNk)
- [Learn AG2 on DeepLearningAI](https://www.youtube.com/watch?v=TBNTH-fwGPE)

--------------------------------------------------------------------------------

WebSurferAgent with Crawl4AITool

Installation

To integrate Crawl4AI with AG2, follow these steps:

Install AG2 with the crawl4ai extra:
```
pip install -U ag2[openai,crawl4ai]
```
If you have been using autogen or ag2, all you need to do is upgrade it using:
pip install -U autogen[openai,crawl4ai]
or
pip install -U ag2[openai,crawl4ai]
as autogen and ag2 are aliases for the same PyPI package.

Set up Playwright:

# Installs Playwright and browsers for all OS
playwright install
# Additional command, mandatory for Linux only
playwright install-deps

For running the code in Jupyter, use nest_asyncio to allow nested event loops.
```
pip install nest_asyncio
```

Once installed, you’re ready to start using the browsing features in AG2.

Imports

import os

import nest_asyncio

from autogen import LLMConfig
from autogen.agentchat import UserProxyAgent
from autogen.agents.experimental import WebSurferAgent

nest_asyncio.apply()

Configure WebSurferAgent with Crawl4AI

Crawl4AI is built on top of LiteLLM and supports the same models as LiteLLM.We had great experience with OpenAI, Anthropic, Gemini and Ollama. However, as of this writing, DeepSeek is encountering some issues.

llm_config = LLMConfig(
    api_type="openai",
    model="gpt-4o-mini",
    api_key=os.environ["OPENAI_API_KEY"],
)

The only difference from the previous example is that the web_tool parameter must be set to crawl4ai in order for the Crawl4AITool to be used. As in previous example, there are two ways to start the chat session:

Recommended: Using the `run` Method

The new run method simplifies the process by eliminating the need for manual UserProxyAgent creation.

✅ Easier setup – No need to manually register tools

# `web_tool` parameter must be set to `crawl4ai` in order for the `Crawl4AITool` to be used.
with llm_config:
    websurfer = WebSurferAgent(name="WebSurfer", web_tool="crawl4ai")

run_response = websurfer.run(
    message="Get info from https://docs.ag2.ai/docs/Home",
    tools=websurfer.tools,
    max_turns=2,
    user_input=False,
)

# Iterate through the chat automatically with console output
run_response.process()

Manual Setup: Using `initiate_chat` Method

This method requires manually creating a UserProxyAgent and registering tools for execution.

⚠️ More setup required
⚠️ Must manually register tools

user_proxy = UserProxyAgent(name="user_proxy", human_input_mode="NEVER")
with llm_config:
    websurfer = WebSurferAgent(name="WebSurfer", web_tool="crawl4ai")

websurfer_tools = websurfer.tools
# WebSurferAgent has a list of tools which are registered for LLM
# We need to register the tools for execution with the UserProxyAgent
for tool in websurfer_tools:
    tool.register_for_execution(user_proxy)

user_proxy.initiate_chat(
    recipient=websurfer,
    message="Get info from https://docs.ag2.ai/docs/Home",
    max_turns=2,
)

Output

user_proxy (to WebSurfer):

Get info from https://docs.ag2.ai/docs/Home

--------------------------------------------------------------------------------

>>>>>>>> USING AUTO REPLY...
WebSurfer (to user_proxy):

***** Suggested tool call (call_UvcLMFR8osM9AsyxVdmzgmBs): crawl4ai *****
Arguments:
{"url":"https://docs.ag2.ai/docs/Home","instruction":"Extract the main sections and any key information such as features, usage guidelines, and any other relevant details."}
*************************************************************************

--------------------------------------------------------------------------------

>>>>>>>> EXECUTING FUNCTION crawl4ai...
Call ID: call_UvcLMFR8osM9AsyxVdmzgmBs
Input arguments: {'url': 'https://docs.ag2.ai/docs/Home', 'instruction': 'Extract the main sections and any key information such as features, usage guidelines, and any other relevant details.'}
[INIT].... → Crawl4AI 0.4.247
[FETCH]... ↓ https://docs.ag2.ai/docs/Home... | Status: True | Time: 1.36s
[SCRAPE].. ◆ Processed https://docs.ag2.ai/docs/Home... | Time: 47ms
INFO:httpx:HTTP Request: GET https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json "HTTP/1.1 200 OK"
/Users/robert/projects/ag2/.venv-crawl4ai/lib/python3.11/site-packages/pydantic/_internal/_config.py:345: UserWarning: Valid config keys have changed in V2:
* 'fields' has been removed
  warnings.warn(message, UserWarning)
14:23:43 - LiteLLM:INFO: utils.py:2825 -
LiteLLM completion() model= gpt-4o-mini; provider = openai
INFO:LiteLLM:
LiteLLM completion() model= gpt-4o-mini; provider = openai
INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
14:23:57 - LiteLLM:INFO: utils.py:1030 - Wrapper: Completed Call, calling success_handler
INFO:LiteLLM:Wrapper: Completed Call, calling success_handler
[EXTRACT]. ■ Completed for https://docs.ag2.ai/docs/Home... | Time: 15.169699541991577s
[COMPLETE] ● https://docs.ag2.ai/docs/Home... | Status: True | Total: 16.59s
user_proxy (to WebSurfer):

***** Response from calling tool (call_UvcLMFR8osM9AsyxVdmzgmBs) *****
[
    {
        "index": 0,
        "tags": [
            "introduction"
        ],
        "content": [
            "# AG2",
            "The Open Source Agent OS"
        ],
        "error": false
    },
    {
        "index": 1,
        "tags": [
            "getting_started"
        ],
        "content": [
            "[Getting Started - 3 Minute](https://docs.ag2.ai/docs/</docs/user-guide/quick-start>)"
        ],
        "error": false
    },
    {
        "index": 2,
        "tags": [
            "key_features"
        ],
        "content": [
            "Key Features",
            "**Multi-Agent Conversation Framework**",
            "AG2 provides multi-agent conversation framework as a high-level abstraction. With this framework, one can conveniently build LLM workflows.",
            "**Easily Build Diverse Applications**",
            "AG2 offers a collection of working systems spanning a wide range of applications from various domains and complexities.",
            "**Enhanced LLM Inference & Optimization**",
            "AG2 supports enhanced LLM inference APIs, which can be used to improve inference performance and reduce cost."
        ],
        "error": false
    },
    {
        "index": 3,
        "tags": [
            "explore_content"
        ],
        "content": [
            "## [Quick StartStart building your AG2 application.](https://docs.ag2.ai/docs/</docs/user-guide/quick-start>)",
            "## [ConceptsWork through the key concepts of AG2 including ConversableAgent, GroupChat, Swarm, and tools.](https://docs.ag2.ai/docs/</docs/user-guide/basic-concepts/installing-ag2>)",
            "## [Advanced ConceptsAdvance to RAG, Code Execution, and more complex GroupChats and Swarms.](https://docs.ag2.ai/docs/</docs/user-guide/advanced-concepts>)",
            "## [Use CasesTry out use case workflows including Customer Service, Travel Planning, and Game Design.](https://docs.ag2.ai/docs/</docs/use-cases/use-cases>)",
            "## [Notebook ExamplesA collection of interactive notebooks across all AG2 topics.](https://docs.ag2.ai/docs/</docs/use-cases/notebooks/Notebooks>)",
            "## [API ReferenceDelve into the AG2 API reference.](https://docs.ag2.ai/docs/</docs/api-reference>)",
            "## [How to ContributeGet involved with AG2 by adding what you need and making the framework even better!](https://docs.ag2.ai/docs/</contributor-guide/contributing>)"
        ],
        "error": false
    },
    {
        "index": 4,
        "tags": [
            "popular_resources"
        ],
        "content": [
            "Popular resources",
            "[Foundation Capital Interview with Dr. Chi Wang](https://docs.ag2.ai/docs/<https:/www.youtube.com/watch?v=RLwyXRVvlNk>)",
            "[Learn AG2 on DeepLearningAI](https://docs.ag2.ai/docs/<https:/www.youtube.com/watch?v=TBNTH-fwGPE>)"
        ],
        "error": false
    }
]
**********************************************************************

--------------------------------------------------------------------------------

>>>>>>>> USING AUTO REPLY...
INFO:httpx:HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
WebSurfer (to user_proxy):

Here is the extracted information from the AG2 documentation website:

### Introduction
- **AG2**: The Open Source Agent OS

### Getting Started
- [Quick Start - 3 Minute Guide](https://docs.ag2.ai/docs/user-guide/quick-start)

### Key Features
1. **Multi-Agent Conversation Framework**: AG2 provides a framework for multi-agent conversations, allowing users to build LLM workflows conveniently.
2. **Easily Build Diverse Applications**: AG2 comes with a collection of systems that cater to a range of applications across various domains and complexities.
3. **Enhanced LLM Inference & Optimization**: The platform supports enhanced LLM inference APIs, improving performance and reducing costs.

### Explore Content
- [Quick Start: Start building your AG2 application.](https://docs.ag2.ai/docs/user-guide/quick-start)
- [Concepts: Learn about key concepts including ConversableAgent, GroupChat, Swarm, and tools.](https://docs.ag2.ai/latest/docs/user-guide/basic-concepts/installing-ag2)
- [Advanced Concepts: Explore topics such as RAG, Code Execution, and complex GroupChats and Swarms.](https://docs.ag2.ai/docs/user-guide/advanced-concepts)
- [Use Cases: Experiment with workflows in areas like Customer Service, Travel Planning, and Game Design.](https://docs.ag2.ai/docs/use-cases/use-cases)
- [Notebook Examples: Access a collection of interactive notebooks covering all AG2 topics.](https://docs.ag2.ai/docs/use-cases/notebooks/Notebooks)
- [API Reference: Detailed exploration of the AG2 API reference.](https://docs.ag2.ai/docs/api-reference)
- [How to Contribute: Guidelines for getting involved with AG2 and enhancing the framework.](https://docs.ag2.ai/latest/docs/contributor-guide/contributing)

### Popular Resources
...

This summary covers the main sections, key features, links to further exploration, and resources available in the documentation.

--------------------------------------------------------------------------------

Conclusion

In this post, we’ve shown how to boost your agents with web browsing abilities using the WebSurferAgent. By using tools like BrowserUseTool and Crawl4AITool, your agents can easily fetch real-time information from the web. This makes your agents more useful and flexible, whether you’re getting data from specific pages or gathering info across different topics. With these tools, AG2 helps you create smarter agents that can navigate the web and bring back the details you need with minimal effort.

Edit this page

Blog

​Introduction

​WebSurferAgent with BrowserUseTool

​Installation

​Imports

​Configure WebSurferAgent with Browser Use

​Recommended: Using the run Method

​Manual Setup: Using initiate_chat Method

​Output

​WebSurferAgent with Crawl4AITool

​Installation

​Imports

​Configure WebSurferAgent with Crawl4AI

​Recommended: Using the run Method

​Manual Setup: Using initiate_chat Method

​Output

​Conclusion

Introduction

WebSurferAgent with BrowserUseTool

Installation

Imports

Configure WebSurferAgent with Browser Use

Recommended: Using the `run` Method

Manual Setup: Using `initiate_chat` Method

Output

WebSurferAgent with Crawl4AITool

Installation

Imports

Configure WebSurferAgent with Crawl4AI

Recommended: Using the `run` Method

Manual Setup: Using `initiate_chat` Method

Output

Conclusion