Status: [LIVE]
Endpoints found:
https://www.quantizelab.dev/api/v1/scan
Source: https://github.com/thecodehaider/Prompt-Injection-Scanner
# 🛡️ Prompt Injection Scanner
### The Open Security Intelligence Layer for LLM Applications
**Detect prompt injections. Audit system prompts. Harden your AI pipelines. — Free.**
[](https://quantizelab.dev/tools/prompt-scanner)
[](https://quantizelab.dev/docs/api)
[](https://quantizelab.dev/pricing)





---
## What Is This?
The **Prompt Injection Scanner** is a production-grade security auditing engine for LLM system prompts. It combines a fast heuristic firewall with a deep AI-powered red-team audit layer — all mapped to the **OWASP LLM Top 10 2025** framework.
This repository supports local heuristic scanning, programmatic SDK operations, and wraps our advanced developer tools: the **Model Context Protocol (MCP) Server** and the **`quantize-brain` Codebase Prompt Auditor CLI**.
---
## 🔌 1. Quantize Lab MCP Server
### What is it?
The Model Context Protocol (MCP) server acts as a native plugin for AI coding agents. Once installed, it equips tools like **Claude Desktop, Cursor, VS Code, and Windsurf** with security tools.