Update README.md

README.md

@@ -11,3 +11,28 @@ license: mit
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# 🔐 LLM Vulnerability Tester  
+
+A simple **Gradio-powered web app** that helps analyze user prompts for potential **security risks in Large Language Models (LLMs)** such as **prompt injection, jailbreaks, data leakage, harmful requests, and hallucination risks**. The tool also suggests a **safer reformulation** of risky prompts.  
+
+---
+
+## 🚀 Features  
+- Detects vulnerabilities in prompts:
+  - Prompt Injection  
+  - Jailbreak / Safety Bypass  
+  - Data Leakage Attempts  
+  - Harmful or Offensive Requests  
+  - Hallucination Risk  
+- Provides **concise risk analysis** with risk levels.  
+- Suggests a **safer reformulation** if risks are detected.  
+- Clean and interactive **Gradio UI**.  
+
+---
+
+## 🛠️ Tech Stack  
+- [Python](https://www.python.org/)  
+- [Gradio](https://www.gradio.app/) – for building the UI  
+- [Groq](https://groq.com/) – LLM API used for analysis and rewriting  
+