AI & Agents MCP Servers for OpenAI Agents SDK

Benchmarks local LLM inference speed (tokens/sec) on your own hardware via MCP tools.

A Model Context Protocol (MCP) server for Lerian portfolio discovery, documentation, learning, SDK examples, live product API

Catholic AI compass: system prompts from Pope Leo XIV's encyclical Magnifica Humanitas.

Generate complete, tested FastMCP 3.x MCP servers from one plain-English sentence.

MEOK AI Labs - neural-health-monitor MCP server extracted from SOV3

Model routing for AI agents: delegate bulk reads & boilerplate to cheap worker models.

Generate images, videos, and audio using 100+ AI models via muapi.ai.

Traces which neurons and attention heads drive a language model's output via MCP tools.

Call your local Ollama or vLLM model over MCP with schema-validated JSON output

Read your recorded AI agent runs, and what changed when you swapped a model or prompt.

MCP server that refines and improves user prompts for AI-optimal readability

Coordinate AI-agent work on a local-first, owner-controlled MCP board

QSP — relay MCP tools to any OpenAI-compatible local LLM (llama.cpp, Ollama, vLLM)

Read-only MCP server for the OPERANT AI operating-agent calibration benchmark.

76% fewer MCP tokens in our standard benchmark. Same upstream call. Exact recovery.

Wraps the slop-eval CLI as a single generic MCP tool for genericness scoring of AI UI output.

Ping 134+ free coding LLMs across 17 providers, rank by latency, pick the fastest.

Agent-only collaboration space: pass the door, then use rooms, DMs and verifiable chains.

MCP Server for ThinkPrompt - Prompt Management System with Multi-Workspace Support

Verifies token/cost savings claimed by AI context-reduction proxies via MCP tools.

Wraps the toolgovern CLI as a single generic MCP tool for agent-tool policy validation.