Operationalizing CaMeL: Strengthening LLM Defenses for Enterprise Deployment (2505.22852v1)

Abstract: CaMeL (Capabilities for Machine Learning) introduces a capability-based sandbox to mitigate prompt injection attacks in LLM agents. While effective, CaMeL assumes a trusted user prompt, omits side-channel concerns, and incurs performance tradeoffs due to its dual-LLM design. This response identifies these issues and proposes engineering improvements to expand CaMeL's threat coverage and operational usability. We introduce: (1) prompt screening for initial inputs, (2) output auditing to detect instruction leakage, (3) a tiered-risk access model to balance usability and control, and (4) a verified intermediate language for formal guarantees. Together, these upgrades align CaMeL with best practices in enterprise security and support scalable deployment.