What is the difference between prompt-based and command hooks?

Prompt-based hooks use LLM reasoning for flexible, context-aware decisions, while command hooks execute bash scripts for fast, deterministic operations like file system checks.

What events can triggers hooks in Claude Code?

Hooks can be triggered by events such as PreToolUse, PostToolUse, Stop, SessionStart, SessionEnd, UserPromptSubmit, and PreCompact.

Can I persist environment variables across a session?

Yes, using a SessionStart hook, you can write export commands to the $CLAUDE_ENV_FILE to persist variables throughout the entire Claude Code session.

How do I block a dangerous command in Claude Code?

You can use a PreToolUse hook with a 'deny' decision output to halt the execution of a tool and provide a reason back to the agent.

How do I ensure my hooks are portable across different environments?

Always use the ${CLAUDE_PLUGIN_ROOT} and ${CLAUDE_PROJECT_DIR} environment variables within your hook configurations to reference paths dynamically.

Claude Code Hook Development

Name: Claude Code Hook Development
Author: SwervoJake

bySwervoJake

0•

Herramientas para Desarrolladores

Simplifies the creation and implementation of event-driven automation hooks to extend and secure Claude Code's capabilities.

Hook Development provides a robust framework for building event-driven automation within Claude Code plugins. By leveraging both prompt-based LLM reasoning and deterministic command-line execution, developers can create sophisticated workflows that validate tool inputs, enforce security policies, and manage project context. This skill is essential for developers looking to build production-grade plugins that require fine-grained control over the interaction lifecycle, from session initialization to final task completion.

Características Principales

01Session management hooks for automated context loading and environment setup

020 GitHub stars

03Granular tool matching using regex and wildcard patterns

04Prompt-based hooks for context-aware LLM validation and reasoning

05Pre-execution validation to block dangerous commands or sensitive path access

06Bash command hooks for fast, deterministic file system and tool checks

Casos de Uso

01Ensuring code quality by running linters or tests via Stop and SubagentStop hooks

02Automating project-specific setup by loading environment variables during SessionStart

03Enforcing security policies by validating file write operations and preventing path traversal

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio hook-development

For use in Claude.ai and ChatGPT

Características Principales

01Session management hooks for automated context loading and environment setup

020 GitHub stars

03Granular tool matching using regex and wildcard patterns

04Prompt-based hooks for context-aware LLM validation and reasoning

05Pre-execution validation to block dangerous commands or sensitive path access

06Bash command hooks for fast, deterministic file system and tool checks

Casos de Uso

01Ensuring code quality by running linters or tests via Stop and SubagentStop hooks

02Automating project-specific setup by loading environment variables during SessionStart

03Enforcing security policies by validating file write operations and preventing path traversal

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add swervojake/claaaude-studio hook-development

For use in Claude.ai and ChatGPT