Lasso Claude Hooks

Claude Hooks

A collection of security and utility hooks for Claude Code. Hooks allow you to extend Claude Code's behavior by running custom scripts at key points during execution.
Research Paper: For detailed analysis of indirect prompt injection vulnerabilities in Claude Code, see: The Hidden Backdoor in Claude Coding Assistant

Available Hooks

🛡️ Prompt Injection Defender

Defense against indirect prompt injection attacks. Scans tool outputs (files, web pages, command results) for injection attempts and warns Claude about suspicious content via PostToolUse hooks.

Quick Start

Option 1: Interactive Installation (Recommended)

If you have this repo added as a Claude Code skill, simply tell Claude:

Claude will handle the entire installation process for you.

Option 2: Install Script

What gets installed

The installer copies hook files to your project and configures Claude Code:

📖 For manual installation and more options, see INSTALLATION.md

Understanding Prompt Injection

The Problem: Indirect Prompt Injection

When Claude Code reads files, fetches web pages, or runs commands, malicious instructions can be hidden in that content:

Without protection, Claude might follow these hidden instructions. The defender scans all tool outputs and warns Claude when suspicious patterns are detected.

Direct vs Indirect Injection

- Direct: Malicious instructions from the user directly (not our focus)
- Indirect: Malicious instructions hidden in content Claude reads (our focus)

Attack Categories

1. System Prompt Forgery - Attempts to override system instructions
2. User Prompt Camouflage - Hidden malicious content in user data
3. Model Behavior Manipulation - Exploiting AI alignment tendencies

Why Pattern-Based Detection?

- Fast: No API calls, instant scanning
-…