Docs / source reviewed
jev-judge-mcp
Give Python MCP agents evidence checks and review routing
How this project uses Jev
- Input
- Caller-supplied claims, evidence, candidates, patches or content to inspect
- Jev decides
- Use Jev’s enumerated answers to assess support, contradiction or missing evidence, returning probabilities and tool-specific policy results
- Code executes
- The caller acts on auto, review or escalate results; missing or invalid answers follow each tool’s review or stopping policy
Evidence and limitations
MIT licensed. A Python rewrite of @jkudish/jev-mcp with an added scoring tool. It judges only supplied material and cannot guarantee truth or force callers to follow advice. Remote calls cost money; HTTP transport is experimental. Accuracy, speed and security were not independently tested.
This project has not been run independently here. Author-reported results are not independently verified results.
Original sources
- https://github.com/PyModel/jev-judge-mcp/blob/7037cf5723bff11d663b838574c5f0e59aa39a49/README.md ↗
- https://github.com/PyModel/jev-judge-mcp/blob/7037cf5723bff11d663b838574c5f0e59aa39a49/LICENSE ↗
- https://github.com/PyModel/jev-judge-mcp/blob/7037cf5723bff11d663b838574c5f0e59aa39a49/src/jev_judge_mcp/providers/typesafe.py ↗
- https://github.com/PyModel/jev-judge-mcp/blob/7037cf5723bff11d663b838574c5f0e59aa39a49/src/jev_judge_mcp/tools/verify.py ↗
- https://github.com/PyModel/jev-judge-mcp/blob/7037cf5723bff11d663b838574c5f0e59aa39a49/src/jev_judge_mcp/policy/actions.py ↗