---
id: 7528
title: "LLM SecRange: Open-Source Penetration Testing Playground for Large Language Models & Vulnerable Agents"
source_domain: github.com
source_url: "https://github.com/gatsby-sec/llm-sec-range"
published_at: "2026-06-20T08:06:45Z"
tags: ["ai-security", "rce", "ctf", "prompt-injection", "red-team", "llm-security", "owasp-llm-top-10", "vulnerable-agent", "training-range"]
phying_url: "https://phying.com/item/7528"
lang: en
---

# LLM SecRange: Open-Source Penetration Testing Playground for Large Language Models & Vulnerable Agents

## Summary

LLM SecRange is a Flask-based local LLM security training range featuring 8 prompt-injection CTF levels, OWASP LLM Top 10 (2025) labs, and a vulnerable ReAct bank agent with switchable target models.

## Why it matters

Offers security researchers and penetration testers an out-of-the-box, comprehensive environment to practice prompt injection, OWASP LLM Top 10 risks, and agent tool-abuse exploitation against real models.

## Intelligence Metadata

- **Sources:** github.com
- **Published:** 2026-06-20T08:06:45Z
- **Original:** https://github.com/gatsby-sec/llm-sec-range
- **Phying News:** https://phying.com/item/7528