Hook Templates for Claude Code

关于

This skill empowers developers to extend Claude Code's functionality by implementing sophisticated hook patterns for session management and tool execution. It offers standardized templates for blocking dangerous commands (Gates), triggering post-task actions like linting or committing (Side Effects), and maintaining persistent workflow states. Whether you need to inject project context at session start or route tasks to specific sub-agents, these templates provide the necessary logic, exit codes, and JSON structures to build a reliable, highly automated development environment tailored to your project's specific needs.

主要功能

• Pre-built Gate patterns for blocking unauthorized or dangerous tool executions based on custom rules.
• Advanced State Manager patterns to track and persist workflow phases across different sessions.
• Context Injector templates to automatically load project metadata and activate services at startup.
• Side Effect templates for automated linting, formatting, and git operations post-execution.
• Orchestration patterns for reliable skill activation and routing to specific plugin agents.
• 0 GitHub stars

使用场景

• Synchronizing Claude's internal state with external APIs or Model Context Protocol (MCP) services for real-time data access.
• Enforcing security protocols by blocking specific shell commands or file modifications until prerequisites are met.
• Automating documentation updates or code commits immediately after Claude successfully modifies the codebase.