AgentThreatBench

Benchmark / dataset

Operationalizes OWASP Top 10 for Agentic Applications (2026) as tests.

Category: Prompt injection & agent security
Insurance exposure: Agent actions
Publisher: Contributed to inspect_evals
Publisher type: Academic
License: MIT
Status: In inspect_evals (active)
Last commit: 2026-10-15
Stars: 704(shared repo star count — UKGovernmentBEIS/inspect_evals backs 19 entries in this catalog)

Wanich Risk Lab, 2026.