Build and trace multi-agent workflows with Microsoft Agent Framework
Compose, orchestrate, and trace production multi-agent workflows with Microsoft Agent Framework.
npx skills add agentskillexchange/skills --skill build-and-trace-multi-agent-workflows-with-microsoft-agent-framework
Use this skill when an operator needs to build or review a production-oriented multi-agent workflow with explicit orchestration, tool calls, handoffs, and tracing. The workflow is to model the agents and tools, define the coordination path, run the workflow against a real task, and inspect OpenTelemetry-style traces or logs before promoting it. Invoke it instead of a plain chat prompt when the task needs repeatable routing, observable state, and a framework boundary across Python or .NET services. Do not use it as a generic SDK listing or as a claim that every application should become a multi-agent system. The scope boundary is concrete workflow construction and trace review using the upstream Microsoft Agent Framework project.
What this skill actually does
Inputs and prerequisites: Python or .NET runtime, Microsoft Agent Framework packages, OpenTelemetry-compatible tracing backend when observability is enabled.
Setup notes: Follow the Microsoft Agent Framework README for the current Python or .NET package install command, then create an agent workflow with tools, handoffs, and tracing configured.
Source and verification boundary: use https://github.com/microsoft/agent-framework as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.