Prompt Injection

2 articles

Browse all articles about Prompt Injection. Explore expert insights and research from the Shinobi team.

Bug of the Week: How a 30-Word PDF Approved Unsafe Products for Shipment

A crafted PDF containing 30 words of plain text bypassed the entire product compliance validation pipeline at a global retail organisation. The LLM read the document, accepted the injected instructions as fact, and approved non-compliant products for shipment.

David Mound

2026-04-15

10 min read

Prompt Injection

LLM Security

Leaking OpenAI's Hidden GPT-5 System Prompt via Context Poisoning

A post-mortem on a critical vulnerability where a "smarter" reasoning model was tricked by a fundamental architectural flaw. What is "Juice: 64"?

Abhishek Gehlot

2025-11-27

12 min read