wtcooper/agent-evals
v0.1.0MIT
Evaluate an AI agent and the features switched on for it: guided reviews of computer use, memory, connectors, code execution and other agent features.
scripts
- agent-feature-review
- lib/evalkit.py
- scripts/_evidence.py
- scripts/templates.py
- scripts/write_review.py
- tests/test_agent_feature_review.py
- agent-misuse-benchmark
- interfaces.py
- lib/evalkit.py
- metrics.py
5 more
- new_experiment.sh
- run_benchmark.py
- scenarios.py
- tests/test_agent_misuse_scaffold.py
- validate_scenarios.py
plugin.json
- name
- agent-evals
- version
- 0.1.0
- description
- Evaluate an AI agent and the features switched on for it: guided reviews of computer use, memory, connectors, code execution and other agent features.
- author
- ai-security-evals contributors
- license
- MIT
plugin.json
{
"$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
"name": "agent-evals",
"version": "0.1.0",
"description": "Evaluate an AI agent and the features switched on for it: guided reviews of computer use, memory, connectors, code execution and other agent features.",
"author": {
"name": "ai-security-evals contributors"
},
"homepage": "https://github.com/wtcooper/ai-security-evals",
"repository": "https://github.com/wtcooper/ai-security-evals",
"license": "MIT",
"keywords": [
"ai-security",
"evaluation",
"agent",
"coding-agent",
"harness"
]
}
What else this package ships
These files come with the package and this site does not publish them. They are listed so you know what is there before you install it.
- 12Python files
- new_experiment.sh
- 12JSON files