How do I share environment variables across a Claude session?

You can use a SessionStart hook to append export commands to the $CLAUDE_ENV_FILE, which persists those variables for all subsequent tools in that session.

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

Prompt-based hooks use natural language instructions for reasoning-heavy decisions, while command hooks execute bash scripts for fast, deterministic operations like file checking.

Can I block specific tool calls using hooks?

Yes, using the PreToolUse hook, you can evaluate the 'tool_input' and return a 'deny' decision to prevent Claude from executing potentially dangerous actions.

Where are hook configurations stored?

Plugin-specific hooks are defined in hooks/hooks.json within the plugin directory, while personal user hooks are configured in .claude/settings.json.

Hook Development for Claude Code

Name: Hook Development for Claude Code
Author: Alongside-cc

byAlongside-cc

•

开发者工具

Implements and configures event-driven automation hooks to validate tool usage and streamline workflows within Claude Code.

This skill provides a comprehensive framework for developing Claude Code hooks, enabling developers to intercept and respond to lifecycle events such as tool calls, session starts, and agent stops. By leveraging both LLM-driven prompt hooks for context-aware reasoning and bash-based command hooks for deterministic checks, users can enforce security policies, automate environment setup, and ensure high-quality outputs. It is an essential toolkit for anyone looking to build robust, safe, and highly customized plugins for the Claude Code CLI.

主要功能

01Deterministic command hooks for fast bash-scripted file and system checks.

02Prompt-based hooks for LLM-driven, context-aware validation and decision making.

031 GitHub stars

04Comprehensive event support including PreToolUse, PostToolUse, SessionStart, and Stop.

05Dual configuration support for both plugin-specific hooks and global user settings.

06Built-in security patterns to prevent path traversal and validate sensitive tool inputs.

使用场景

01Automating project context loading by setting environment variables during SessionStart.

02Implementing quality gates that prevent the agent from finishing until tests or builds pass.

03Enforcing security policies by blocking dangerous bash commands or unauthorized file modifications.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add alongside-cc/5-years-teaching-azure-claude-code hook-development

For use in Claude.ai and ChatGPT

Download Skill