{"version":"2.1.284","anchor":"consent-prompt-for-claude-written-mods","canonical_anchor":"consent-prompt-for-claude-written-mods","heading":"Consent prompt for loading Claude-written mods gains an explanation option","tier":"notice","area":"Mods","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.284\/e\/consent-prompt-for-claude-written-mods","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.284","markdown":"### Consent prompt for loading Claude-written mods gains an explanation option\n\nThe prompt to load mods, hook code Claude wrote, now puts \"? How does this work\" first and can explain up to 3 times\n\n**Unclear.** It is not clear what makes this question appear or whether a feature flag controls it.\n\n**What**\n\nMods are hook modules that Claude wrote. A hook is code that runs automatically at set moments. Before mods from a folder are loaded, Claude Code asks \"Enable hot reloading for this session?\". Hot reloading means picking up changes without restarting.\n\n- The options are now \"? How does this work\", then \"Enable for this session\", then \"Not now\". Before, there were two options, with \"Not now\" first.\n\n- Choosing \"? How does this work\" asks again with a longer explanation, such as \"A mod is code Claude wrote; it runs with your permissions.\"\n\n- The explanation can be asked for up to 3 times. After that, the answer counts as \"Not now\".\n\n- Your answer is recorded as `plugin_dev_mods`.\n\n- Mods load straight away, or \"when this turn ends\" if Claude is in the middle of a turn.\n\n**Why**\n\nThis prompt is an approval step before code Claude wrote runs as hooks with your permissions. You can now read what that means before you agree.\n\n- Area: Mods\n- Tier: You'll notice\n- Useful: 4\/5\n- Signal: 4\/5"}