Hooks

Scripts that fire automatically at lifecycle events for deterministic control over AI behavior.

Why it matters

Hooks give deterministic control over non-deterministic AI. They ensure certain checks always happen.

In practice

We use Claude Code hooks to enforce quality gates: pre-commit hooks run linting, post-tool hooks validate changes.

Related terms

Claude Code

Anthropic's agentic coding tool that runs in the terminal, understands codebases, and executes tasks autonomously.

Guardrails

Rules and constraints that prevent an agent from taking harmful or unauthorized actions.

Skill

A reusable instruction set that teaches an agent a specific workflow.

Back to glossary