Layered defense design for a coding copilot deployment against tool-use hijacking attacks, using structured function-call-only interface on GPT-4.1.
Layered defense design for a coding copilot deployment against prompt leaking attacks attacks, using structured function-call-only interface on o1.
Layered defense design for a coding copilot deployment against system prompt extraction attacks, using hash-based prompt pinning on Gemini 2.0 Flash.
Layered defense design for a coding copilot deployment against payload smuggling in code blocks attacks, using hash-based prompt pinning on Claude 3.5 Sonnet.
Layered defense design for a coding copilot deployment against Unicode homoglyph attack attacks, using output schema enforcement on o1-mini.
Layered defense design for a coding copilot deployment against invisible text injection (zero-width chars) attacks, using output schema enforcement on DeepSeek-V3.
Layered defense design for a coding copilot deployment against memory poisoning attack attacks, using spotlighting (delimiter marking) on Claude 3.7 Sonnet.
Layered defense design for a coding copilot deployment against recursive self-instruction attacks, using spotlighting (delimiter marking) on o3.
Layered defense design for a coding copilot deployment against direct prompt injection attacks, using input sanitization on Llama 3.3 70B.
Layered defense design for a coding copilot deployment against jailbreak prefix attacks, using input sanitization on Claude 4 Sonnet.
Layered defense design for a coding copilot deployment against encoded payloads (base64/ROT13) attacks, using output content filter on Grok 3.
Layered defense design for a coding copilot deployment against tool-use hijacking attacks, using output content filter on Llama 3.1 405B.