What are Claude Code hooks?

Hooks are automated scripts, prompts, or subagents that execute at specific points in the Claude Code lifecycle, such as before a tool is used or after a task is completed.

What is the difference between a command hook and an agent hook?

Command hooks run standard shell scripts, while agent hooks spawn a specialized subagent with tool access to perform complex multi-step validations.

Where are hook configurations stored?

Hooks can be defined globally in ~/.claude/settings.json or project-specifically in .claude/settings.json or skill YAML frontmatter.

Do hooks slow down Claude Code?

While synchronous hooks wait for completion to ensure control, you can use 'async: true' for background tasks that don't block the main conversation flow.

Can hooks block dangerous commands?

Yes, hooks like PreToolUse can inspect command inputs and return a 'deny' decision to prevent destructive actions like unauthorized file deletions.

Hooks Automation for Claude Code

Name: Hooks Automation for Claude Code
Author: Aznatkoiny

byAznatkoiny

0•

Productividad y Flujo de Trabajo

Automates Claude Code workflows by triggering custom shell commands, LLM prompts, and subagents at specific lifecycle events.

Hooks Automation provides a comprehensive system for managing deterministic control over Claude Code's behavior through the use of 14 distinct lifecycle events. By configuring hooks for events like SessionStart, PreToolUse, and TaskCompleted, developers can enforce strict project rules, block destructive commands, and automate repetitive tasks such as linting, testing, or desktop notifications. The skill supports command-based, prompt-based, and agent-based handlers, allowing for everything from simple shell scripts to complex multi-turn subagent validations, ensuring that critical project standards are maintained automatically rather than relying on LLM discretion.

Características Principales

01Built-in security controls to block dangerous commands or protect files

02Deterministic execution of shell commands, prompts, or subagents

03Access to 14 lifecycle events including PreToolUse and TaskCompleted

04Regex-based matcher patterns for granular tool and event filtering

050 GitHub stars

06Support for async background tasks and quality gate enforcement

Casos de Uso

01Injecting project-specific context or documentation at the start of a session

02Preventing accidental execution of destructive shell commands like 'rm -rf'

03Automatically running linters or test suites after code modifications

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add aznatkoiny/claude-dev-toolkit hooks-automation

For use in Claude.ai and ChatGPT

Download Skill