Copilot

Harness

The runtime in Copilot Studio that sits between what you build and the AI model, deciding when to call the model, what to send it and which tools to use.

The harness sits between what you build and the model Your agent or workflow on the left, the AI model on the right, and the harness in the middle deciding when to call the model and which tools to run. YOU BUILD RUNTIME MODEL Your agent or workflow Harness the runtime in between AI model reasons and writes Tools run by the harness

Whatever you build in Copilot Studio runs on a harness. You design the agent or workflow, the model supplies the reasoning, and the harness sits in between. Microsoft offers three: the GitHub Copilot harness for reasoning-heavy, multi-step processes, the standard harness for rule-based agents and structured, repeatable conversations, and the Copilot chat harness for extending Copilot Chat with your organisation's knowledge.

The choice affects how the work gets done, what you can build out of the box and how it is billed. Billing is the difference to watch. The standard harness starts charging Copilot Credits after you publish; the GitHub Copilot harness charges from the moment you start building, previews and tests included.

Often confused with: The model. The model is the AI that reasons and writes; the harness is the runtime around it that calls the model, passes it the right components and runs the tools.

See it in a lesson: Copilot Studio: build and automate

Further reading

← All terms

Rollo AI AcademyWant your whole team speaking this language?Knowing the words is a start. Rollo Academy gets teams genuinely working with AI, through hands-on, in-company training built around your own work.Discover Rollo Academy →