Pinned Loading

1. ai-detects-if-cve-was-zero-day ai-detects-if-cve-was-zero-day Public

   Multi-agent AI system using GPT-4o, DeepSeek v3, and Llama 3.3 to detect if CVE vulnerabilities were exploited as zero-days. Analyzes 50 verified CVEs with 85%+ accuracy using forensic evidence ex…

   Python 17 5

2. LLM-Number-Convergence-Study LLM-Number-Convergence-Study Public

   Investigating Cognitive Biases and Convergence Patterns in Large Language Models' Pseudo-Random Number Selection

   Jupyter Notebook

3. agents-claude-code agents-claude-code Public

   🚀 100 hyper-specialized AI agents for Claude Code - Transform Claude into your personal tech army with experts in React, AWS, Kubernetes, ML, Security & more

   3

4. llms-write-vulnerable-code-they-can-detect llms-write-vulnerable-code-they-can-detect Public

   Experimental framework proving LLMs can detect security vulnerabilities with 97% accuracy but still generate the same vulnerable code when asked. Tests SQL injection, XSS, and command injection ac…

   Python

5. testing-if-ai-abandons-truth-when-pressured testing-if-ai-abandons-truth-when-pressured Public

   Experimental framework testing if LLMs abandon correct answers (like 'Ottawa is Canada's capital') when subjected to psychological manipulation tactics used on humans: gaslighting, authority pressu…

   Python