InjecAgent

Evals

Benchmark / dataset

Indirect prompt injection on tool-integrated agents.

Category
Prompt injection & agent security
Insurance exposure
Prompt injection
Publisher
UIUC (Kang Lab)
Publisher type
Academic
License
MIT
Status
Stale (>12mo)
Last commit
2024-07-02
Stars
176

View source on GitHub →

Wanich Risk Lab, 2026.