← ClaudeAtlas

claude-hooks-settingslisted

Claude Code settings and hooks are created, audited, or debugged across CLAUDE.md/rules loading, custom agents, tool permissions, output styles, workflows, setting sources, safe/bare modes, and plugin customizations.
Xopoko/plug-n-skills · ★ 11 · AI & Automation · score 77
Install: claude install-skill Xopoko/plug-n-skills
# Claude Hooks And Settings Bundled commands use `$PLUGIN_ROOT` (`$env:PLUGIN_ROOT` in PowerShell; same path suffix) for the plugin root. Set it once: use the host's plugin-root variable when defined (Claude Code: `PLUGIN_ROOT="$CLAUDE_PLUGIN_ROOT"`), otherwise the absolute path of this plugin's root directory. Use this skill for Claude Code settings and customization work: settings JSON, hooks, CLAUDE.md or rules loading, custom agents, tool allow/deny rules, output styles, workflows, setting sources, plugin-provided customizations, safe-mode comparisons, and bare-mode explicit context. Read `$PLUGIN_ROOT/references/hooks-settings.md` for hook lifecycle, setting-source boundaries, and safety details. ## First Move Inspect the target files and scope before editing: - user settings: machine-wide, do not commit; - project settings: shareable only when the repo intentionally wants them; - local project settings: machine-local, normally ignored; - plugin settings: bundled with a plugin and active only when enabled; - managed policy settings: admin-controlled and may override local changes. Use CLI evidence when behavior differs from expectation: ```bash python3 "$PLUGIN_ROOT/scripts/claude_code_inspector.py" --commands auto-mode --json claude auto-mode config claude auto-mode defaults ``` ## Hook Design Use hooks only when deterministic lifecycle automation is needed. Prefer narrow events and matchers: - tool gating or audit: `PreToolUse`, `PermissionRequest`, `Permiss