{"version":"2.1.314","anchor":"hook-collection-gains-a-skill-scoped-prompt-hook-filter","canonical_anchor":"hook-collection-gains-a-skill-scoped-prompt-hook-filter","heading":"Hooks from inactive skills can now be limited to prompt hooks","tier":"internal","area":"Hooks","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.314\/e\/hook-collection-gains-a-skill-scoped-prompt-hook-filter","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.314","markdown":"### Hooks from inactive skills can now be limited to prompt hooks\n\nWhen gathering hooks, Claude Code can now keep only prompt-type hooks from skills that are not active, instead of keeping or dropping all of them\n\n**Unclear.** It is not stated what turns on this filtering or what the further check on each hook tests.\n\n**What**\n\nA hook is an action Claude Code runs automatically at a set moment. A skill is a packaged set of instructions that can bring its own hooks. Before, hooks from skills that were not currently held were either all dropped or all kept. Claude Code now has a third option: keep only the hooks of the prompt type, meaning hooks that are a piece of text for the model, and only those that pass a further check.\n\n**Why**\n\nThis looks like groundwork for skills that define their own prompt hooks.\n\n- Area: Hooks\n- Tier: Under the hood\n- Useful: 2\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}