fleet-parallel

Batch code modifications across multiple files run in parallel by launching independent Copilot agents for each sub-task via Fleet mode. You define specific sub-tasks and provide per-agent briefs to apply a single intention to dozens of files simultaneously.

Yes, you can generate unit tests for multiple files simultaneously by using the /fleet command. This applies parallel AI agents to bulk test generation, eliminating the sequential waiting time that blocks developer workflow.

Copilot Fleet mode supports branch isolation for mass refactoring by using Git worktrees. This ensures branch safety while applying a single refactoring intention, such as converting a component pattern, across many files at once.

To track progress for parallel AI agents during bulk documentation generation, you optionally record each sub-task in the session SQL database. This tracks each independent AI reasoning item as it processes.

The best way to apply a single refactoring pattern to dozens of files at once is using Fleet mode. It parallelizes batch refactoring by assigning independent Copilot agents to each file, requiring only sub-task definitions and per-agent briefs.

Yes, you need to define sub-tasks manually and provide per-agent briefs to parallelize test generation. This ensures each independent AI agent receives specific instructions for generating accurate unit tests across files.