SAI LabsBook a demo

Safety IndexEvaluationsChatInject on AgentDojo

ChatInject on AgentDojo

Adversarial robustness · Prompt injectionCounts in the indexBuilt, not yet run

Whether an agent stays on task when a tool result is forged to look like new conversation turns written in the model's own chat format.

Metric
% survival. Higher is better.
Dataset
arxiv.org/abs/2509.22830 · AgentDojo suites · unverified (code: github hwanchang00)
Grader
programmatic (AgentDojo) (the paper used same)
Status
Built, not yet run
Replicability
public

Results

Built and tested; the first full runs are pending.