Blocks AI agent tool calls acting on untrusted data before they cause damage.

Evaluating provenance-gated tool calls as a prompt injection defense on AgentDojo. Reproducible runs, per-case analysis, and published results.

Policy-as-code enforcement and observability for MCP tool calls. Wraps AI agent sessions with cryptographic integrity checks, argument-level CEL policies, and a full audit trail.

Generate YARA rules automatically from positive and negative examples. For PII detection, secret scanning, and prompt injection.

Live PoC: MCP attacks that compromise AI agents mid-session and how to block them in a few lines of code.

Reproducible security benchmarking for the Deconvolute SDK and AI system integrity against adversarial attacks.

A demonstration of RAG poisoning attacks using dormant documentation injections.