
jailbreaks
Curated collection of LLM jailbreak prompts and bypass techniques, documenting adversarial inputs that circumvent AI model safety guardrails.
adversarial-attackai-securitycurated-resources+1
11.0k

Curated collection of LLM jailbreak prompts and bypass techniques, documenting adversarial inputs that circumvent AI model safety guardrails.

Curated collection of 200+ cybersecurity interview questions and answers covering Red Team, Blue Team, Web Security, Incident Response, and network…

Curated collection of free, interactive AI-powered training materials for security awareness education, with community contributions and practical…

Curated collection of tools, bookmarks, and guides for open-source intelligence (OSINT) gathering, reconnaissance, and information gathering.

Draw.io libraries for threat modeling diagrams