Send Research to a Subagent

You ask how the auth layer handles token refresh. It reads forty files into your main conversation, and now the actual implementation runs on a stuffed context, exactly when you need it sharpest.

Ask for a subagent instead: "use subagents to investigate how our authentication handles token refresh, and whether we have OAuth utilities to reuse."

The exploration happens in a separate context window. What comes back is a summary.

All that reading stays off your books. The main session stays clean for the work that actually matters.