PyPI page
Home page
Author:
Simon Sure
Summary:
AgentHarm SecurityClaim for superred: faithful port of the harmful test_public split (176 behaviors) against the general inspect-agent target, reusing upstream inspect_evals.agentharm tools/grading/judges. 8 per-category subclaims + total.
Latest version:
0.1.0
Required dependencies:
huggingface-hub
|
inspect-ai
|
inspect-evals
|
numpy
|
superred
|
superred-target-inspect-agent
Optional dependencies:
pytest
|
pytest-asyncio
|
pytest-cov
Downloads last day:
0
Downloads last week:
2
Downloads last month:
14
7-day smoothing