io.github.ipezygj/evalgateai

evalgate

Verified · today

Statistical checks an agent runs before trusting an AI eval number (is #1 real, judge bias, more).

Install

claude mcp add evalgate -- uvx eval-integrity

Our take

A small MCP server for sanity-checking AI evaluation results, with tools aimed at judge bias, ranking stability, and related statistical pitfalls. It is most useful for teams running agent or model evals who want a quick guardrail before trusting a headline score. The scope is clear and technically relevant, but with minimal visible adoption it reads as early-stage and worth validating on real workloads before depending on it.

reviewed by hand · 2026-07-24

Something wrong or dead here? Report it

Verification record

last verified
today
github stars
1
last commit
today
archived
no
license
MIT
in registry since
2026-07-23

github.com/ipezygj/evalgate

More in ai

browse all →
claude-flowai

claude-flow

Verified · 15 days ago

AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.

hand-reviewed63k starschecked 15 days ago

tldrawai

tldraw

Verified · 16 days ago

Draw and visually collaborate with your agents on tldraw's canvas.

hand-reviewed49k starschecked 16 days ago

codebase-memory-mcpai

Codebase Memory

Verified · 12 days ago

Codebase knowledge graph for AI agents — 159 languages, sub-ms queries, 99% fewer tokens.

hand-reviewed30k stars5.2k dl/wkchecked 12 days ago

openmetadata-mcpai

OpenMetadata

Verified · 8 days ago

Official OpenMetadata MCP: governed context and business semantics for AI assistants and agents.

hand-reviewed14k starschecked 8 days ago

praisonaiai

PraisonAI

Verified · 12 days ago

AI Agents Framework with Self Reflection and MCP support

hand-reviewed8.4k starschecked 12 days ago

strataai

strata

Verified · 5 days ago

MCP server for progressive tool usage at any scale (see https://klavis.ai)

hand-reviewed5.8k starschecked 5 days ago