GitHub Copilot
I spend most of my day inside VS Code, and GitHub Copilot has become the tool I reach for constantly: writing IaC, debugging KQL queries, scaffolding pipelines, reviewing pull requests. The default experience, where Copilot picks up whatever context it happens to find, leaves a lot on the table.
This section is about steering the model deliberately. Not prompt engineering in the "write a better sentence" sense. I mean the structural tools GitHub ships that let you shape the system prompt, inject persistent context, and build reusable agent workflows that understand your codebase.
Why this is not just tips and tricks
The built-in Copilot system prompt is generic by design. It does not know your team uses Terraform with a specific module structure, or that your Python projects always use ruff and pytest, or that your pipelines run on Airflow with particular DAG conventions. Every session I start without custom instructions burns tokens re-explaining context the tool should already have.
Custom instructions, skills, and custom agents are the mechanisms GitHub provides to fix that. I think of them as layers of specificity, and getting the layering right is what makes Copilot feel like a team member instead of autocomplete.