agents-hooks
Claude Code Hooks + Codex Notifications
Use this skill when hook behavior is the main concern: Claude lifecycle hooks, runtime preflight, guardrails, async verification, worktree lifecycle checks, subagent coordination, or Codex notification callbacks.
Claude and Codex are not equivalent here. Claude has a broad, stable hook system. Codex has two surfaces: the long-stable notification surface (notify + tui.notifications), and a lifecycle-hooks system (hooks.json; see developers.openai.com/codex/hooks) whose events, handler types, and trust rules change between releases and have had firing-reliability gaps. Both event sets grow, so look up the current list in each runtime's hooks reference instead of trusting a count. This skill owns the Codex hook facts as lookup steps in references/hook-templates.md §Codex. Prefer notify for anything that must be dependable, and confirm a Codex hook fires on your runtime before relying on it.