PyPI page
Home page
Author:
None
Summary:
Production-ready evaluation framework for AI agents — 58 metrics (25 native + 33 Harness Config) across 7 evaluation gates: goal achievement, behavioral integrity, reliability, performance, security, multi-agent coordination, and observability
Latest version:
0.9.1
Required dependencies:
anthropic
|
numpy
|
openai
|
pandas
|
python-dotenv
Optional dependencies:
anthropic
|
arize-phoenix
|
autogen-agentchat
|
autogen-core
|
build
|
crewai
|
datasets
|
deepeval
|
dspy-ai
|
fastapi
|
jinja2
|
kiwipiepy
|
langchain
|
langchain-anthropic
|
langchain-core
|
langchain-openai
|
langgraph
|
mlflow
|
mypy
|
openai
|
openpyxl
|
opentelemetry-exporter-otlp-proto-http
|
opentelemetry-sdk
|
pdfplumber
|
pre-commit
|
pyarrow
|
pyautogen
|
pydantic-ai
|
pytest
|
pytest-asyncio
|
pytest-cov
|
python-multipart
|
ragas
|
ruff
|
sentence-transformers
|
twine
|
uvicorn
|
wandb
Downloads last day:
17
Downloads last week:
101
Downloads last month:
1,353