Skip to content

AgentHarm

Named by 2 kept items in this workspace.

Search every workspace for AgentHarm. It matches the name as written, so it can find other things that share it.

The items that name it 2

  • AI Security Briefing
    repo2026-07-19

    Public agent-security dataset used to build the cross-source benchmark for validating generalization

  • agent-egress-bench
    repo

    Model-alignment benchmark for whether the LLM refuses harmful instructions

All technologies in this workspace.