Multi-Agent-Custom-Automation-Engine-Solution-Accelerator  by microsoft

Solution accelerator for multi-agent automation MVP

Created 9 months ago
536 stars

Top 59.4% on SourcePulse

GitHubView on GitHub
Project Summary

This solution accelerator provides a foundational, AI-driven system for orchestrating multiple specialized AI agents to automate complex tasks. It's designed for developers and businesses looking to build Minimum Viable Products (MVPs) or reference applications for multi-agent workflows, offering a significant head start in leveraging generative AI for business efficiencies.

How It Works

The system utilizes the AutoGen framework to manage a team of AI agents, coordinated by a manager agent. A FastAPI backend handles user requests, which are then processed by a Planner agent to create an execution strategy. This plan is delegated to specialized agents (e.g., HR, Procurement) for task completion. State is maintained across sessions using Azure Cosmos DB, and human feedback can be incorporated into the workflow. This approach enables parallel processing and specialized task execution, improving efficiency and consistency.

Quick Start & Requirements

  • Deployment: One-click Azure deployment via a provided button, or deployment via GitHub Codespaces, VS Code Dev Containers, or local environment using Azure Developer CLI (azd).
  • Prerequisites: An Azure subscription with permissions to create resources. Access to Azure OpenAI Service, Azure AI Search, and Azure Cosmos DB is required, with specific attention to Azure OpenAI quota availability.
  • Setup Time: Deployment to Azure is estimated at 7-10 minutes.
  • Documentation: User Story, Quick Deploy, Supporting Documentation.

Highlighted Details

  • Leverages Microsoft's AutoGen framework for agent orchestration.
  • Employs a FastAPI backend for API requests and Azure Cosmos DB for state management.
  • Demonstrates use cases like employee onboarding, procurement, and press release drafting.
  • Designed for customization, allowing integration of custom data and agent capabilities.

Maintenance & Community

The project is maintained by Microsoft Research's AI Frontiers Lab. Further information on the AutoGen framework can be found via a linked blog post.

Licensing & Compatibility

The solution accelerator itself is not explicitly licensed in the README. However, it relies on Azure services (Azure OpenAI, Azure Cosmos DB, Azure Container App, Azure Container Registry), which are subject to Microsoft's terms of service and pricing. Use of these services requires compliance with applicable Microsoft Product Terms and export laws.

Limitations & Caveats

This is a solution accelerator and has not been tested as an end-to-end, reliable production application. Users are strongly encouraged to perform their own performance and safety evaluations before deploying to production. The system is not designed for high-risk use cases where failure could lead to death, serious injury, or environmental damage.

Health Check
Last Commit

14 hours ago

Responsiveness

1 day

Pull Requests (30d)
124
Issues (30d)
3
Star History
53 stars in the last 30 days

Explore Similar Projects

Starred by Luis Capelo Luis Capelo(Cofounder of Lightning AI), Addy Osmani Addy Osmani(Head of Chrome Developer Experience at Google), and
23 more.

goose by block

0.8%
20k
Open-source AI agent for automating complex engineering tasks
Created 1 year ago
Updated 21 hours ago
Starred by Elie Bursztein Elie Bursztein(Cybersecurity Lead at Google DeepMind), Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), and
7 more.

SuperAGI by TransformerOptimus

0.1%
17k
Open-source framework for autonomous AI agent development
Created 2 years ago
Updated 7 months ago
Starred by Wes McKinney Wes McKinney(Author of Pandas), Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), and
22 more.

autogen by microsoft

0.5%
50k
Agentic framework for multi-agent AI applications
Created 2 years ago
Updated 19 hours ago
Feedback? Help us improve.