How to evaluate AI agents in Microsoft Copilot Studio
Agent Evaluation in Copilot Studio helps makers move from early optimism to grounded confidence as agents grow in complexity and impact.
Copilot
Learn how agentic AI can help automate and execute business processes for different teams.
Agent Evaluation in Copilot Studio helps makers move from early optimism to grounded confidence as agents grow in complexity and impact.
Learn six key capabilities organizations are using to scale Copilot Studio agent adoption in 2026—plus practical considerations for enterprise deployment.
The Microsoft Copilot Studio extension for Visual Studio Code is generally available, so you can build and manage Copilot Studio agents from the IDE you already use.
In this edition of our monthly roundup, we’re highlighting a few of our biggest updates from Microsoft Ignite 2025 and walking through new capabilities available today.
Today we’re excited to bring OpenAI’s GPT-5.2 to Microsoft 365 Copilot and Microsoft Copilot Studio.
Claude Opus 4.5 is now in Copilot Studio, offering sharper reasoning, better long‑context, and stronger multi‑step performance for your agents.
Explore new Microsoft Copilot Studio updates to shape agent behavior, enforce organizational standards, and support agentic business transformation.
Today, we are making GPT-5.1 available in Microsoft Copilot Studio, alongside OpenAI’s release. This is available as an experimental model for U.S. customers in early release cycle Power Platform environments. The GPT-5.1 series brings improved adaptability in thinking time in both chat and reasoning.
In this edition of our monthly roundup, we’re recapping new features released in Microsoft Copilot Studio in October 2025.
Today, we’re bringing AI-powered building to employees across the organization, with new agents for Microsoft 365 Copilot customers in the Frontier program: App Builder and Workflows.
Automated agent testing is now built into Copilot Studio—evaluate performance, improve quality, and scale confidently with Agent Evaluation.
Introducing the Copilot Credit Pre-Purchase Plan (P3)—a flexible way for organizations to purchase Copilot Credits and scale their agent initiatives with confidence.