#evals

Smarter Agent

Minimal, self-checking, simulation-driven improvement for AI agents

Researchers have investigated whether AI labs are deliberately training models to excel at a humorous, unscientific benchmark involving peli...

Califa Cards

Califa Cards: A Comprehensive Skill Card Standard and Testing Framework

Scorable MCP

MCP for Scorable Evaluation Platform

Scenarios

A portable, model-agnostic test suite of declarative YAML scenarios for evaluating AI assistant security against prompt-injection and data-e...

Prefect MCP Server

Prefect MCP server

My Pi

Composable Pi coding agent with MCP, LSP, agent chains, prompt presets, and local eval telemetry

Mastra

Mastra is the modern TypeScript framework for AI-powered applications and agents.