- Category
- Developer Tools
- Rank
- No. 1634Tools index
- Pricing
- Open Source
- Type
- TOOL
- Use case
- Model & Agent Evaluation
- Interfaces
- CLI
- GitHub
- 17 stars
- Latest release
- v3.5.2
- Date
About
An offline, deterministic evaluation framework for AI agents. It checks OpenTelemetry GenAI traces, A2A AgentCards and MCP tool lists against 165 rule-based tests, so it can flag unauthorized tool calls, prompt injections, unearned success claims and PII leaks without an LLM judge. Because results do not depend on a model, runs are repeatable and can be added to CI.
Tags
agent-evalsopentelemetrytestingmcpai-safetypython
Tech Stack
PythonDocker
Media
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.