All projects

tenbin

simota/tenbin

MCP server and agent skill for the TypeSafe AI System One API (Jev): decompose a judgment into Choice / Score / Noul questions, lint them, measure on labelled data, and put calibrated thresholds in code

TypeScript
Stars
4
Forks
2

Why it is listed

Tenbin combines an agent skill and MCP tools for designing and measuring Jev questions; production code calls the TypeSafe SDK directly, not the MCP.

Related projects

  • jev-native-mcp

    Review public code and documents with Jev in Codex and Claude Code: ranking, batch screening, claim checks, spend limits, and evaluation tools.

    Shared topics: classification, claude-code, mcp

  • ripwire-broker

    Task-oriented, budgeted MCP code context over Ripwire: three tools, hooks for Claude Code and Codex, optional relevance scoring and per-workspace memory via TypeSafe Jev, optional local-model notes

    Shared topics: claude-code, mcp

  • jev-dotnet

    Unofficial .NET SDK for TypeSafe AI's Jev System One model. Typed Choice/Score/Noul judgments with calibrated confidence, zero dependencies.

    Shared topics: classification, confidence