JevOss
by Community
Evaluation tools for Jev-compatible decision models, covering accuracy, calibration, hidden instructions, padding and consistency.
OSS
JevOss
Added 16 Oct 2026
Overview
JevOss provides evaluation tools for decision models that are compatible with the Jev framework. It measures model performance across accuracy, calibration, hidden instruction resistance, padding behavior, and consistency. Written in Python, it is available as a community project on GitHub.
Best for
Best for
Developers working with Jev decision models who need lightweight evaluation checks.
Use cases
- Benchmark Jev-compatible decision models
- Test model calibration and consistency
- Check resistance to hidden instructions and padding
Notes
2 stars on GitHub. Last updated 2026-10-15. Licensed Apache-2.0.
Pros
- Covers multiple evaluation dimensions in one tool
- Simple Python-based tooling for Jev models
- Open source and free to use
Cons
- Very low community adoption (2 stars)
- Limited to Jev-compatible models
- No evidence of active maintenance or documentation
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.