The Raccoon LAM Tool enables AI agents to browse and interact with the web to perform tasks like: - Executing simple and complex web tasks and workflows across multiple sites - Web navigation and form submission - Data extraction from websites - Online research and information gathering ## Key Features - **Web Browsing and Web Tasks**: Automated navigation of web pages and completion of user defined tasks - **Data Extraction**: Structured data extraction from websites - **Schema Definition**: Define the structure of data you want extracted ## Capabilities - Search and browse websites - Fill out forms and navigate UI elements - Extract structured data based on defined schemas - Handle multistep tasks across websites ## Schemas and Deepsearch - Schemas are used only when you want to extract information from the web. - Deepsearch is only used if answering the query involves gathering data from multiple sources and detailed reports. - Schemas can be used alongside deepsearch. - Schemas should not be used when the user query is about performing actions/task rather than data collection Args: query: The input query string for the request response_schema: The expected schema for the response (optional) app_url: The entrypoint URL for the web agent (optional) chat_history: Chat history as list of messages (optional) max_count: Maximum number of results (default: 1) stream: Whether to stream responses (default: True) mode: Mode of execution ("default" or "deepsearch") advanced: Advanced configuration options ctx: The context Returns: The LAM results as a formatted string

Return a sample LAM query to demonstrate the API functionality.