This third installment of Cloud Wars’ coverage from last week’s Microsoft Build 2025 event delivers details on Azure AI Foundry enhancements, additional new AI development functionality in Foundry, Copilot Studio and GitHub, and new AI agents.
In previous reports, we’ve detailed Copilot Tuning and multi-agent orchestration functions, as well as a major multi-agent interoperability push largely focused on Model Context Protocol or MCP.
A wide range of new Azure AI Foundry platform updates advance developer functionality with support for additional models, an agent service, multi-agent orchestration, and more. In terms of new development functionality, Microsoft rolled out major enhancements to Copilot Studio and the Github development platform focused on AI-centric development.
When it comes to new AI agents, the company’s rollouts included task-specific agents for managing documents, customer engagement, and sales leads.
Azure AI Foundry Advances
In Azure AI Foundry, the company has expanded developer choice to now include over 1,900 partner- and Microsoft-hosted AI models while AI Foundry provides underlying data integration, model customization, and governance services.
Newly supported models include the Grok 3 and Grok 3 mini models from xAI, Flux Pro 1.1 from Black Forest, and Sora via Azure OpenAI in Foundry Models. Over 10,000 open-source models from Hugging Face are now available in Foundry Models.
An enhanced, smarter model system within Azure AI Foundry makes it easier to choose the right model for each task. A new Model Router automatically chooses the best model for a given prompt while maximizing quality and minimizing costs. Among new tools in Azure AI Foundry is a Model Leaderboard that ranks the top-performing AI models for different categories and tasks.
AI Agent & Copilot Summit is an AI-first event to define opportunities, impact, and outcomes with Microsoft Copilot and agents. Building on its 2025 success, the 2026 event takes place March 17-19 in San Diego. Get more details.
The newly available Azure AI Foundry Agent Service gives developers tools to orchestrate multiple specialized agents so they handle complex tasks. The service manages infrastructure and orchestration; it comes with ready-to-use templates, actions, and connectors to more than 1,400 enterprise data sources such as SharePoint, Microsoft Fabric, and third-party systems.
New Azure AI Foundry Observability features deliver metrics on performance, quality, cost and safety, all incorporated alongside detailed tracing in a dashboard user interface.
The new Microsoft Entra Agent ID provides functionality so agents that developers create in Azure AI Foundry or Copilot Studio are automatically assigned unique identities in an Entra directory, helping enterprises securely manage agents from the start and avoid “agent sprawl” or security blind spots.
Other Azure AI Foundry enhancements (see the full list) include multi-agent orchestration allowing agents to call each other as tools, passing tasks between specialized agents to solve complex problems. This includes support for Agent-to-Agent (A2A) communication to enable different agents to exchange information and coordinate tasks, as well as MCP.
The final Azure AI Foundry update we’ll cover here is agentic retrieval in Azure AI Search, a new query engine designed to process complex questions, using conversational context and an embedded large language model to break complex queries down into sub-queries to maximize answer relevance.
AI Agent & Copilot Summit is an AI-first event to define opportunities, impact, and outcomes with Microsoft Copilot and agents. Building on its 2025 success, the 2026 event takes place March 17-19 in San Diego. Get more details.
Copilot Studio Enhancements + New Agents
The Build announcements included a range of new Copilot Studio features and a new set of agents spanning specific use cases and business functions.
A new Tools tab in Copilot Studio serves as a centralized place to create and manage reusable functionality across agents and take actions in external systems.
Copilot Studio has these tool types:
- MCP, which allows users to connect with knowledge servers and data sources directly within Copilot Studio
- Agent flows, which are used to automate workflows
- Computer use, which allows agents to navigate and interact with web and desktop applications
- Custom connectors and REST APIs, which support connections to third-party systems that aren’t available as prebuilt connectors
- Prompts for creating AI-powered instructions for smarter agents, flows, and apps.
Three new agents were also introduced at Build:
- Document Processor Agent for automating document workflows, which monitors an email inbox for attachments, extracts information from incoming files, and exports structured data to a target system
- Customer Brief Agent pulls from business data to generate timely, relevant executive briefs before client meetings
- Lead Manager Agent acts as a top-of-funnel assistant, autonomously processing and responding to inbound leads
GitHub Advances
Microsoft’s GitHub announcements included a GitHub Copilot coding agent and new updates to GitHub Models.
The company said GitHub Copilot is evolving from an in-editor assistant to an agentic AI partner on the strength of a new asynchronous coding agent. Microsoft is also adding prompt management, lightweight evaluations and enterprise controls to GitHub Models so teams can experiment with various AI models without leaving GitHub.
Microsoft is also open-sourcing GitHub Copilot Chat in VS Code. The AI-powered capabilities from GitHub Copilot extensions will now be part of the same open-source repository that drives the underlying GitHub technology that’s used by over 150 million developers.
Closing Thoughts
With the AI development- and agent-focused introductions detailed in this report, alongside previous new products and features for multi-agent orchestration and interoperability coming from Build 2025, Microsoft continues to move the expand the breadth and depth of its agentic AI strategy to support customers’ initiatives for building, operating, and extracting the most value from AI agents.
The rapid delivery of new AI agent technologies and services from Microsoft is applying pressure to competitors and partners to sustain the pace and join Microsoft in driving industrywide innovation.
Ask Cloud Wars AI Agent about this analysis
