From the Team
Blog
AI security insights, tutorials, and research from the QbitDojo team.
5 articles published
How to Test LLM Applications for Prompt Injection
A practical, five-phase methodology for testing LLM apps and AI agents for prompt injection: system prompt extraction, direct override, indirect injection, output and tool abuse, and defense evaluation. With success-rate scoring and a CI-ready test suite.
What Is the OWASP Top 10 for LLM Applications? A Developer's Breakdown
A plain-English breakdown of the 2025 OWASP Top 10 for LLM Applications: what each risk means, a real-world example of each, what changed since 2023, and how to practice every category hands-on.
How to Prevent Prompt Injection in Production LLM Apps
A practical guide to defending production LLM applications and AI agents against prompt injection: seven layered defenses, from input checks and spotlighting to least privilege, CaMeL-style isolation and continuous red teaming.
Top 5 Prompt Injection Techniques Developers Must Know
The five prompt injection techniques behind real AI incidents, from Bing's leaked system prompt to zero-click attacks on AI assistants. How each one works, why it is dangerous, and how to defend against it.
What Is LLM Prompt Security? A Practical Guide (2026)
What LLM prompt security is, why it matters now that AI agents can act on our behalf, how attackers exploit AI systems, and the defenses that actually help. Real incidents, plain English, hands-on labs.