ActionEngine leverages the llm model and the embedding model to output code from the prompt and the html page. Args: llm (`LLMPredic

CommandCenter allows you to launch a gradio demo powered by selenium and the ActionEngine Args: actionEngine (`BaseActionEngine`):

TestActionEngine removes any querying and returns a default code - is used to quickly test software Args: dummy_code: (`str`)

Helper class to visit AST nodes and extract variables assigned in the code.

Adapter for HtmlRetriever to be compatible with the llama index BaseRetriever

Class for Vision-based WorldModel

Mainly for benchmarks, do not use it as the performances are not up to par with the other retrievers

Abstract class for ActionEngine

Abstract class for WorldModel

Class for Vision-based WorldModel

LiteLLM supports 100+ LLM APIs. See the complete list: https://docs.litellm.ai/docs/providers Since the api_key would depend on which provide