Action Orchestration

In this example, we will demonstrate how to use ActionWeaver to orchestrate the design of hierarchies and chains of action.

Keep in mind that orchestration is limited to a single action invoked per API call. If multiple actions are invoked, same set of actions will be available to LLM again

import os
import openai
from openai import OpenAI

openai.api_key = os.getenv("OPENAI_API_KEY")

Patch OpenAI client

from actionweaver.llms import wrap

openai_client = wrap(OpenAI())

Define function you want model to invoke

from actionweaver import action
from typing import List

@action(name="FileHandler")
def handle_file(instruction: str) -> str:
    """
    Handles user instructions related to file operations. Put every context in the instruction only!

    Args:
        instruction (str): The user's instruction about file handling.

    Returns:
        str: The response to the user's question.
    """
    print ("Handling file...")
    return instruction


@action(name="ListFiles")
def list_all_files_in_repo(repo_path: str ='.') -> List:
    """
    Lists all the files in the given repository.

    :param repo_path: Path to the repository. Defaults to the current directory.
    :return: List of file paths.
    """
    print (f"List all files in {repo_path}...")

    file_list = []
    for root, _, files in os.walk(repo_path):
        for file in files:
            file_list.append(os.path.join(root, file))
        break
    return file_list

@action(name="ReadFile")
def read_from_file(file_path: str) -> str:
    """
    Reads the content of a file and returns it as a string.

    :param file_path: The path to the file that needs to be read.
    :return: A string containing the content of the file.
    """
    print (f"Read file from {file_path}")
    with open(file_path, 'r') as file:
        content = file.read()
    return f"The file content: \n{content}"

Build a hierarchy of actions

Once FileHandler invoked by LLM, LLM can access actions from ListFiles, ReadFile, or reply with a message.

FileHandler
  |
  |-- ListFiles
  |
  |-- ReadFile

response = openai_client.create(
  model="gpt-3.5-turbo",
  messages=[{"role": "user", "content": "list all files"}],
  actions = [handle_file],
  orch = {
        handle_file.name: [read_from_file, list_all_files_in_repo]
  },
  stream=False,
)

Handling file...
List all files in ....