Repolode
Submit
microsoft

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

An Azure-based reference accelerator where multiple specialized AI agents plan and execute user-specified organizational tasks, built on Microsoft Agent Framework and Azure AI Foundry.

View on GitHub
Stars
890
Forks
845
Open issues
29
Last commit
6 hours ago

What is it?

This project is a solution accelerator for building AI-driven orchestration systems. A user describes a task, and a group of specialized AI agents coordinates to plan, execute, and validate it. It is aimed at complex organizational workflows that span several departments, such as release planning, employee onboarding, or contract review.

The stack is Microsoft Agent Framework and Azure AI Foundry for the agents and models, Azure Cosmos DB for storing metadata and processing results, Azure AI Search for vector-based data search, Azure Container Apps for the web frontend, and Azure Container Registry for images. Deployment is done with Azure Developer CLI (azd) and Bicep templates, and the project can be opened in GitHub Codespaces or Dev Containers.

The project is a proof of concept and reference application rather than a finished product. Its documentation notes that it supports English input and output only, and that generated output should be verified before use. Running it incurs Azure usage costs that vary by region and usage.

Key features

  • 01Coordinates multiple specialized AI agents to plan, execute, and validate user-specified tasks.
  • 02Uses Microsoft Agent Framework and Azure AI Foundry for agent orchestration and models.
  • 03Stores metadata and processing results in Azure Cosmos DB.
  • 04Provides vector-based data search through Azure AI Search.
  • 05Deploys to Azure with Azure Developer CLI (azd) and Bicep templates.

Use cases

  • Prototyping multi-agent workflows for product marketing release planning.
  • Automating employee onboarding steps that span HR, IT, and compliance.
  • Reviewing RFPs or contracts for risks and compliance with a team of agents.
  • Using as a reference architecture for building agent systems on Azure.

Quick start

  1. 1

    Check Azure OpenAI quota availability before deploying

  2. 2

    Install Azure Developer CLI version 1.18.0 or higher

  3. 3

    Follow the deployment guide to deploy the solution to your Azure subscription

  4. 4

    To tear down the resources when finished, run azd down

    azd down

Things to know

  • Deployment requires an Azure subscription and incurs usage-based costs.
  • Requires Azure Developer CLI 1.18.0 or higher and Bicep CLI 0.33.0 or higher for local deployment.
  • Some tenants with security restrictions may break the app, requiring the WAF-supported version.

Topics

#multi-agent#ai-agents#azure#orchestration#llm#bicep#automation#ai-azd-templates#azd-templates

FAQ

What is Multi-Agent-Custom-Automation-Engine-Solution-Accelerator?
An Azure-based reference accelerator where multiple specialized AI agents plan and execute user-specified organizational tasks, built on Microsoft Agent Framework and Azure AI Foundry. This project is a solution accelerator for building AI-driven orchestration systems. A user describes a task, and a group of specialized AI agents coordinates to plan, execute, and validate it. It is aimed at complex organizational workflows that span several departments, such as release planning, employee onboarding, or contract review.
Is Multi-Agent-Custom-Automation-Engine-Solution-Accelerator free to use? What license does it have?
Multi-Agent-Custom-Automation-Engine-Solution-Accelerator is open source under the MIT license. Check the license terms before using it in commercial products.
Is Multi-Agent-Custom-Automation-Engine-Solution-Accelerator still maintained?
Actively maintained. The last commit to Multi-Agent-Custom-Automation-Engine-Solution-Accelerator was on Oct 21, 2026.
What can I use Multi-Agent-Custom-Automation-Engine-Solution-Accelerator for?
Prototyping multi-agent workflows for product marketing release planning. Automating employee onboarding steps that span HR, IT, and compliance. Reviewing RFPs or contracts for risks and compliance with a team of agents. Using as a reference architecture for building agent systems on Azure.
How do I install Multi-Agent-Custom-Automation-Engine-Solution-Accelerator?
Check Azure OpenAI quota availability before deploying Install Azure Developer CLI version 1.18.0 or higher Follow the deployment guide to deploy the solution to your Azure subscription To tear down the resources when finished, run azd down: azd down
Where can I get Multi-Agent-Custom-Automation-Engine-Solution-Accelerator?
The source code and installation instructions are on GitHub: https://github.com/microsoft/Multi-Agent-Custom-Automation-Engine-Solution-Accelerator

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects