LLM-as-judge guardrail for pi · from yapp

Judgementalist — judge Sherlock contests faster

LLM-as-judge evaluation engine — score agent outputs with structured rubrics

CLI and TypeScript library that has multiple LLMs draft, refine, and judge one final answer.

Atbash safety judge exposed as a standalone MCP server

A package to automatically check/judge answers against quizbowl answerlines.

Deterministic, local-first WASI and WASIX compiler and judge library powered by Wasmer.

The `Framebuffer Picker` tool using GPU frame picking technology, it can judge whether a certain model is clicked at the pixel level, and return information such as the material and mesh of the picked entity.

Open source AI evaluation framework — LLM-as-judge + assertion-based evals for any AI app. CLI + MCP server.

Three-layer MCP tool server (judge, suite, gate) for agent-eval-harness

Contract-synthesis + match-judge engine for ggui's handshake. Synthesizes or repairs a conforming DataContract from an agent's draft, judges blueprint-match candidates for reuse, and validates contract structure + novelty — the primitives composed by deci

Deterministic design engine for Urbicon UI — the zero-dependency design linter, manifest parser and quality rubric that power the design loop (validate · context · judge).

Simple Judger ================ A simple npm module to judge C/C++ program.(Linux only) using [EasySandbox](https://github.com/daveho/EasySandbox) for safety judge

Agentic terminal investigator for C/C++ memory leaks — MCP-driven static + dynamic (Valgrind/ASan/LSan) analysis with an LLM judge.

Shared types, Zod schemas, and framework-free C/C++ memory-leak analysis (heuristic + consensus judge, report renderers) for the cleak toolchain.

Thin, typed JavaScript/TypeScript client for the Omega-API geometric judge.

GitHub Actions entrypoints for judge-java

Goal-driven autonomous continuation plugin for pi-dashboard — surfaces the @ricoyudog/pi-goal-hermes judge loop as a live session-card chip plus set/pause/resume/clear controls.

typeof-plus is a small tool to judge the data type of variables

Three-role AI deliberation CLI: engineer proposes, genius challenges, judge points the next question.