bytes of thoughtby Daniel Tucker
Hooks in Claude Code run shell commands at lifecycle events like PreToolUse and Stop. Learn to configure them and build a hook that blocks secret exposure.
Skills define the approach. Subagents execute it. A framework for choosing between them, with a real example combining both in a single Claude Code command.
Claude Code subagents run in isolated contexts and return results to your session. How they work, how to spawn them, and patterns for parallel execution.
Skills are reusable instruction files that tell Claude Code how to handle a class of task — how they work, how to install them, and when to reach for one.
See all tags.