>_ The Manifest

A declarative agent that can only chat is a demo. One that can read a document, run real code, and draw a chart is a tool people actually keep open. The gap between those two is capabilities, and it is smaller than you think. Watch the walkthrough, then skim the recap below.

▶ Configuring your Copilot Chat declarative agent capabilities on the Microsoft 365 & Power Platform Community channel. Open on YouTube.

What you’ll see

Capabilities are the built-in powers you switch on in the manifest to move an agent beyond plain conversation. This walkthrough turns a basic agent into a context-aware assistant by layering three of them.

  • Embedded knowledge so the agent answers from documents you ship with it
  • Code interpreter so it can run Python to analyze data and produce results
  • Image generation so it can create visuals on demand
  • Where each one goes in the manifest and how they combine in a single agent

The takeaway: capabilities are configuration, not code. You declare what the agent can do, and the platform brings the runtime.

Resources

Have questions or want to share what you're building? Connect with me on LinkedIn or check out more on The Manifest.