The GitHub Copilot harness in Copilot Studio is now the Microsoft Copilot harness
See how the Microsoft Copilot harness is helping Copilot Studio makers build more capable, reliable agents for complex business challenges.
Copilot
Discover what’s coming to Microsoft Ignite, Nov 17-20, 2026. Register now.
Agentic AI
-
-
Microsoft named a Leader in two evaluations of next-generation app development platforms
In this evolving market, we’re proud that Microsoft has been named a Leader in two analyst evaluations, each with its own market definition and assessment criteria. -
Introducing the customer conversation harness in Microsoft Copilot Studio
Build customer-facing agents in Copilot Studio with a new conversation harness that connects voice, chat, business knowledge, and contact data. -
Delegate shared business processes to agents that are more persistent and proactive with Microsoft Copilot Studio
Delegate shared business processes to persistent agents that work across your organization, carry context forward, and improve over time. -
New and improved: Build apps, extend agents, and transform business processes in Microsoft Copilot Studio
Discover what’s new in Copilot Studio, September 2026: new capabilities that bring apps, workflows, and agents together to transform business processes. -
Build where you want, run with confidence: Now Microsoft hosts and manages the code created by Copilot
Meet Microsoft Copilot Managed Runtime, an enterprise platform that helps teams build and run apps with security, governance, and IT control. -
Build apps in Copilot Cowork and Copilot Studio
Introducing app building in Copilot Cowork and Copilot Studio. -
Extending agentic Microsoft Dynamics 365 Sales with MCP
MCP extends Dynamics 365 Sales agents with external data, tools, and actions, helping connect sales processes across business systems. -
Building reliable voice agents: A practical guide
Learn how to design AI voice agents that handle real conversations, stay reliable under pressure, and scale from prototype to production. -
Who evaluates the evaluators? The data science behind agent evals
An inside look at the data science and evaluation systems helping teams improve agent quality at scale in Copilot Studio. -
Frontier Tuning: Teaching AI to work the way you do
Frontier Tuning introduces a new way to tune AI agents around your organization’s workflows, internal knowledge, and compliance safeguards. -
Mistral joins Copilot Studio’s growing lineup of model providers
Copilot Studio adds Mistral Medium 3.5, expanding model choice with in‑region data control, strong multilingual performance, and admin governance.