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 code formatting or validation.

Where are Claude Code hook configurations stored?

Hooks can be configured globally in ~/.claude/settings.json, per-project in .claude/settings.json, or locally in .claude/settings.local.json. Claude merges matching hooks from all three locations.

How do hooks receive data from Claude Code?

Hooks receive information as a JSON object through standard input (stdin). This JSON includes session details, the tool being used, and the specific inputs or outputs for that tool.

Can a hook block Claude from running a dangerous command?

Yes. By using a PreToolUse hook and exiting with status code 2, the hook will block the operation and display your custom error message to Claude.

Claude Code Hooks Configuration

Name: Claude Code Hooks Configuration
Author: Ven0m0

byVen0m0

0•

Herramientas para Desarrolladores

Automates Claude Code workflows and enforces development best practices through lifecycle-based shell command hooks.

This skill provides expert guidance for configuring and developing Claude Code hooks to create deterministic, automated workflows. It covers the entire hook lifecycle—including SessionStart, PreToolUse, and PostToolUse—enabling developers to inject context, auto-format code, block dangerous commands, and send notifications. By utilizing these standardized patterns and schemas, users can ensure consistent agent behavior and integrate Claude Code deeply into their existing development environments and local automation pipelines.

Características Principales

01Comprehensive mapping of lifecycle events like SessionStart, PreToolUse, and Stop

02Detailed Input/Output schemas for JSON-based hook communication

03Pre-built patterns for auto-formatting, command blocking, and audit logging

04Best practices for secure shell scripting and performance optimization

050 GitHub stars

06Multi-level configuration guidance for user, project, and local settings

Casos de Uso

01Enforcing security policies by blocking destructive shell commands before execution

02Injecting Git status or environment-specific context at the start of every session

03Automatically running formatters like Prettier or Ruff after Claude edits a file

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add ven0m0/claude-config hooks-configuration

For use in Claude.ai and ChatGPT

Download Skill