"Allowing an AI agent to ask user questions during a task" matching MCP servers:

Enables an AI agent to boot and control an Android emulator — tapping, typing, scrolling, inspecting screens, reading logs and installing APKs — and to record those interactions as replayable, CI-ready test flows. Actions are captured by element text or id rather than coordinates, so the saved tests survive layout changes and rerun without the agent.

Exposes tools over stdio that let an AI agent create disposable Docker sandboxes to run commands and stream output, read and write files, and take snapshots, roll them back, fork them, list, and destroy them. It tells the model to snapshot before risky changes so failed attempts can be undone or explored in parallel on self-hosted infrastructure.

Gives a coding agent its own Docker-hosted Linux desktop with XFCE, Chromium, Blender, LibreOffice, GIMP and a sudo shell, exposing tools for screenshots, mouse and keyboard input, window focus, shell commands, file read/write, opening URLs or GUI apps, and headless Blender renders. The agent can watch and drive that desktop live from a browser while working on the shared project folder.

MCP server that lets an AI agent manage Proxmox VE in natural language, with policy-driven security, read-only mode, two-step confirmations, and audit logging.

Dockerfile Generator AI - MCP server providing AI-powered tools and automation by MEOK AI Labs

Enables users to list GKE clusters, describe virtual machines, and retrieve billing information through MCP read tools, without applying production changes.

An MCP server for managing home lab VMs and PNETLab topologies, enabling AI agents to power VMs, handle snapshots, and verify device configurations with explicit allowlists and confirmation gating.

Docker Helper MCP server that provides tools for automating Docker tasks such as generating Dockerfiles, parsing docker-compose files, and optimizing images.

A VMware ESXi/vCenter management server based on MCP (Machine Control Protocol), providing simple REST API interfaces for virtual machine management.

MCP server for the Agent Platform API, enabling management of sandbox environments through MCP protocol. Exposes tools to interact with sandbox lifecycle operations such as start, stop, and status checks.

MCP server for managing Proxmox VE hosts, providing VM health reports, alerts, and a chat interface via terminal or Cursor.

Provides MCP tools for managing disposable Docker sandboxes that let AI agents safely execute commands in isolated, ephemeral environments.

MCP server for automating VLP lab VM operations, exposing VM management tools to AI agents like Cursor and Claude Code.

Enables AI-assisted DevOps by exposing scoped, role-based tools to inspect and remediate a Docker app stack, with audit logging and allow-listed commands for safe incident response.

MCP server that gives AI agents isolated, named Linux computers with stable IDs and lifecycle control. Supports multiple backends like Docker, Fly Machines, Modal, and E2B, with a dashboard for managing sandboxes.

MCP server to give every agent an ephemeral Linux sandboxes for executing shell commands.

MCP server for securely discovering, pricing, renting, connecting, and releasing GPU compute instances from AI Galaxy with budget checks and two-phase approval.

Enables AI clients to manage Huawei eNSP simulated network devices, execute CLI commands, apply configurations, and perform network diagnostics through natural language, with topology upload and visualization via a web UI.

Enables a computer-use agent to see and drive one isolated, sandboxed Linux desktop by exposing Anthropic's computer toolset actions alongside launching apps and listing windows, with screenshots scaled to a 1280-pixel longest side. It runs as a stdio MCP server scoped to a single screen and authenticated by an agent token.

Enables interaction with VMware SDDC Manager through natural language, exposing tools to list domains, check health, manage hosts, and more.

Enables AI agents to manage and interact with Kasm Workspaces containerized desktop infrastructure, providing tools for session management, command execution, file operations, and user management via the Model Context Protocol.

A Python-based server that enables AI assistants to interact with an Unraid server through the official Unraid GraphQL API, providing read-only access to system information, Docker containers, VMs, storage, and more.

Enables users to query Azure HPC/AI Kubernetes clusters for GPU node information and InfiniBand topology details through kubectl commands. Provides tools to list GPU pool nodes with their status and retrieve network topology labels for high-performance computing workloads.

An MCP server that gives LLM agents full API access to a VMware Cloud Foundation 9.1 estate, exposing 8,931 operations through eight tools with automatic authentication, async task tracking, and write safety.

A Model Context Protocol (MCP) server that interfaces with the Apache YuniKorn Scheduler, allowing AI agents to observe and reason about Kubernetes batch workload resource management, queue hierarchies, and application states.

MCP server for managing Proxmox VE clusters — provision VMs and containers, manage snapshots and backups, execute commands, browse storage, and monitor resources through natural language

Enables an assistant to control a Tandy Color Computer 3 emulated in MAME, allowing typing, disk operations, screen viewing, and state management.

Enables programmatic control of a PNETLab network-emulation lab via natural language, allowing users to build topologies, wire links, push device configs, and boot the lab through an LLM agent.

Exposes a HashiCorp Nomad cluster as MCP tools for AI agents, enabling read-only inspection of jobs, allocations, nodes, and deployments, with optional write operations like stopping jobs, scaling task groups, restarting allocations, and registering jobs.

A Model Context Protocol server that enables AI clients to interact with virtual Ubuntu desktops, allowing them to browse the web, run code, and control instances through mouse/keyboard actions and bash commands.