Input Sanitization for Agentic Systems: What Actually Works

Jun 18, 2026 · 6:30 PM · 100 Queens Quay East, Toronto

Input Sanitization for Agentic Systems: What Actually Works

KC Udonsi on the production sanitization layer that sits between untrusted input and the model.

KU

KC Udonsi

Venue

100 Queens Quay East, Toronto

Speaker

KU

DC416 co-organizer. AI defense lead. Works at Stan (DC416's recurring venue sponsor). Co-presenter for May 21 on the defensive half of AI in security.

About

Agentic systems amplify every classic LLM safety problem. A prompt injection isn't a jailbreak anymore, it's remote code execution by way of your assistant, or an intellectual property or data leak. A PII leak isn't a compliance footnote, it's training data for a vendor's next model. And as agents start reading tool outputs, retrieved documents, and other agents' messages, the trusted-input boundary disappears entirely.

This talk walks through the design of a production sanitization layer that sits between untrusted input and the model, regardless of whether that input comes from a user, a tool, or another agent.

What we cover

  • Why generic guardrails fail. Regex gets bypassed in seconds. Bracketed PII redaction like [NAME_1] actively provokes hallucinations. Single-classifier approaches miss paraphrased attacks.
  • A layered detection model: heuristics, fine-tuned classifiers, semantic drift, and LLM-as-judge. When each pays for itself and when it doesn't.
  • Context-preserving pseudonymization: replacing PII with structurally valid fakes (real names, reserved IPs, 555-phones) instead of placeholders, and why this keeps downstream reasoning intact.
  • Integration trade-offs: transparent proxy vs SDK hook vs sidecar gRPC. Latency budgets, blast radius, and the operational cost of each.

You leave with a concrete reference architecture, the failure modes we hit in production, and the numbers behind why some "obvious" defenses make things worse.