MCP server for mobile app automation: verify, control, and debug iOS, Android, TV, and desktop apps
3,797 MCP servers
Securely run LLM-generated code in isolated containers across 7 languages and 3 container backends.
Unified AI router with 160+ providers, RTK+Caveman compression, auto fallback, MCP/A2A, desktop, PWA, and OpenAI-compatible APIs.
Lint AI agent context files (CLAUDE.md, AGENTS.md, etc.) against your actual codebase
Agent-native TypeScript framework for MCP servers.
The open-source governance layer for AI agents. Byte-identical run receipts, 40+ adapters, air-gap.
TypeScript AI SDK with 24+ LLM providers behind one consistent API. MCP-native (connect any MCP server), voice TTS/STT/realtime, RAG
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
2,500+ scientific tools for AI scientists: life science, research, literature, and more.
Model Context Protocol server for Forest Admin with OAuth authentication
MCP Server for Z.AI - A Model Context Protocol server that provides AI capabilities
Unified CLI and MCP server for Google NotebookLM
ChatGPT (Codex CLI) and Claude (Claude Code) discuss until they agree, on your own subscriptions.
MCP server for dossier automation standard - enables LLMs to discover, verify, and execute dossiers
Runtime control, proof, and fleet intelligence for MCP-compatible AI agents.
ContextForge AI Gateway — an AI gateway, registry, and proxy for MCP, A2A, and REST/gRPC APIs. Exposes a unified control plane with
A Model Context Protocol (MCP) server for DataHub
Local-first MCP server for LLM output verification — risk signals, injection and PII checks.
Route AI agent tasks to the best MCP server and LLM, scored on 132+ real benchmark executions.
An intelligent Model Context Protocol (MCP) Server that acts as an AI model advisor. Compare LLM and media pricing, features, and quality
Cut your Claude Code bill, measured on your own sessions. Profile and slim any LLM conversation.
AI & Agents MCP servers on mcprush, filed under one heading so the shelf can be read in one pass.
Every server here names the publisher behind it and whether that publisher has been verified, the runtime it uses, how it authenticates, the tool surface it exposes and the price model it is sold under — and where a security scan has been run against it, what the scan found. The filters on the left narrow it further and the sort at the top decides what comes first.
Nearly every server in this catalogue is connected straight to its publisher: the listing gives the publisher’s own address or start command, and mcprush is not in the path, so nothing is routed, metered or billed here. The exception is a server delivered through the mcprush gateway — there one key covers everything installed against your account, the calls are metered where you can see them and the ceiling on what they may cost is yours to set, and installing it is a line of the mcprush CLI: npx mcprush@latest add, followed by the server’s id. The listing says which of the two it is before you install anything.