# Guides Inspect and rewrite commands Match, question, walk, and rewrite the Bash command an agent is about to run. Register an external transcript Under the hood The dispatch pipeline, verdict precedence, the trust model, and the resident daemon. What captain-hook does not do The blunt list: what capt-hook deliberately does not do. LLM hooks, signals & the transcript Put a model's judgment in a hook -- LLM gates and nudges, signal scoring, and transcript queries. Packs & configuration Use the packs you have, ship your own in a Claude plugin, and tune hook settings. Writing hooks Primitives, conditions, and style rules -- the vocabulary every hook is built from. Session reviewer How repeated corrections become hooks: the reviewer, its thresholds, and the PR it opens. State & workflows Typed session state, durable state, and multi-step workflow gates. Test hooks inline Inline tests={…}, the test runner, and replaying recorded sessions. Troubleshooting Symptom-by-symptom fixes for hooks that don't fire, fire wrong, or fire twice.