This folder implements the CodeAct idea (paper, tweet) that consolidates LLM agents’ actions into a unified code action space for both simplicity and performance (see paper for more details).

The conceptual idea is illustrated below. At each turn, the agent can:

1. Converse: Communicate with humans in natural language to ask for clarification, confirmation, etc.
2. CodeAct: Choose to perform the task by executing code
   • Execute any valid Linux bash command
   • Execute any valid Python code with an interactive Python interpreter. This is simulated through bash command, see plugin system below for more details.

To make the CodeAct agent more powerful with only access to bash action space, CodeAct agent leverages OpenDevin's plugin system:

• Jupyter plugin: for IPython execution via bash command
• SWE-agent tool plugin: Powerful bash command line tools for software development tasks introduced by swe-agent.

Example of CodeActAgent with gpt-4-turbo-2024-04-09 performing a data science task (linear regression)

[] Support web-browsing [] Complete the workflow for CodeAct agent to submit Github PRs