Where are hook configurations stored?

Hooks can be configured at multiple levels: user-level in ~/.claude/settings.json, project-level in .claude/settings.json, or locally in .claude/settings.local.json.

How do I inject context into a session automatically?

Use the SessionStart hook to execute a script that returns a JSON object containing an 'additionalContext' field. This information will be automatically added to the Claude session on startup.

What are Claude Code hooks?

Hooks are user-defined shell commands that execute at specific points in the Claude Code lifecycle, providing guaranteed, deterministic execution for tasks like formatting, logging, or safety checks.

Can I prevent Claude from running dangerous commands?

Yes, by using the PreToolUse hook with a script that inspects tool input and returns exit code 2, you can effectively block and provide a reason for stopping destructive operations.

Claude Code Hooks Configuration

Name: Claude Code Hooks Configuration
Author: laurigates

bylaurigates

•

Productividad y Flujo de Trabajo

Automates development workflows and enforces best practices by configuring deterministic shell command hooks within the Claude Code lifecycle.

Acerca de

The Claude Code Hooks Configuration skill provides expert guidance for implementing lifecycle hooks that execute shell commands at specific points during Claude Code operations. It enables developers to automate repetitive tasks like code formatting, block dangerous command execution, inject project-specific context at session start, and set up desktop notifications. By moving from Claude's probabilistic tool usage to deterministic hook execution, teams can ensure consistent behavior and rigorous safety standards across their development environments.

Características Principales

Input/output schema guidance for custom hook development
Deterministic execution of shell commands and scripts
Full lifecycle hook support from SessionStart to SessionEnd
Tool-specific matching for precise automation control
Pre-built patterns for safety, formatting, and notifications
3 GitHub stars

Casos de Uso

Blocking destructive root commands like 'rm -rf' before they execute
Automatically formatting Python or TypeScript files after tool-based edits
Injecting real-time git status as context at the start of every session

Acerca de

Características Principales

Input/output schema guidance for custom hook development
Deterministic execution of shell commands and scripts
Full lifecycle hook support from SessionStart to SessionEnd
Tool-specific matching for precise automation control
Pre-built patterns for safety, formatting, and notifications
3 GitHub stars

Casos de Uso

Blocking destructive root commands like 'rm -rf' before they execute
Automatically formatting Python or TypeScript files after tool-based edits
Injecting real-time git status as context at the start of every session