lorenzofavaro / agentic-workflow-orchestrator

A modular Python orchestrator for agentic workflows, enabling multi-step task planning, model routing, debate, judging, and verification.

Home Page:https://medium.com/@lorenzofavaro/stop-wasting-money-on-llms-a-smart-orchestrator-for-cost-efficient-ai-30fba76d193e

Repository from Github https://github.comlorenzofavaro/agentic-workflow-orchestratorRepository from Github https://github.comlorenzofavaro/agentic-workflow-orchestrator

Agentic Workflow Orchestrator

Overview

This repository presents an Agentic Workflow Orchestrator, a modular framework for coordinating multiple AI agents in collaborative decision-making pipelines. The system enables the integration of diverse agent roles (planning, judging, debating, verifying, and routing) and supports interaction with various AI providers.

Architecture

Architecture

Getting Started

Prerequisites

  • Python 3.11+
  • API keys for supported AI providers (e.g., OpenAI, Google)

Setup

  1. Clone the repository:

    git clone <repo-url>
    cd agentic-workflow-orchestrator
  2. Create and activate a virtual environment:

    python3 -m venv .venv
    source .venv/bin/activate
  3. Install dependencies:

    pip3 install -r requirements.txt
  4. Configure environment variables:

  • Example environment file is provided as .env.example in the app/ directory.
  • To configure your environment:
    1. Copy the example files to their active counterparts:

      cp app/.env.example app/.env
    2. Edit it to set your API keys. For example:

      OPENAI_API_KEY=your-key
      GOOGLE_API_KEY=your-key
  1. Run the orchestrator
    cd app
    python3 main.py

About

A modular Python orchestrator for agentic workflows, enabling multi-step task planning, model routing, debate, judging, and verification.

https://medium.com/@lorenzofavaro/stop-wasting-money-on-llms-a-smart-orchestrator-for-cost-efficient-ai-30fba76d193e

License:MIT License


Languages

Language:Python 100.0%