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

Lets Claude Code conduct long, structured interviews via a keyboard-driven local web form, with unlimited questions and options, recommendations, ranking, previews, and live follow-up questions.

An autonomous MCP server that uses the Claude Code CLI to solve coding problems without permission prompts.

Enables AI agents to create, submit answers, and query profile questions from the White Rabbit Profile Questions API. Supports various question types and batch operations.

Enables real-time user intervention for MCP agents through a web UI, allowing users to review context and provide feedback when AI agents drift from intent, keeping them on track.

Enables AI assistants to query hundreds of models across OpenAI, Google, and OpenRouter, with model discovery, deep-research jobs, and image generation.

Provides read-only tools to search and retrieve timestamped transcript evidence from the Pinkbike Podcast, enabling bike research questions to be answered with cited source passages.

Enables ChatGPT Developer Mode to answer questions using operator-defined instructions, with tools to set and retrieve those instructions via a local Streamable HTTP MCP server.

Central user-state database MCP server with durable profile facts and live virtual sensors (stress, room intensity) with configurable smoothing, readable by any agent or smart-home service.

Enables coding agents to request guarded code and architecture reasoning from multiple AI models through a single MCP server, with modes for direct answers, parallel councils, ordered chains, and structured debates, plus reusable context and audit tracing.

MCP server for token-user-system wallet operations, supporting balance, trend, transactions, coupons, and corporate transfer info.

MCP server enabling Claude Code to capture screenshots of Windows applications through WSL2, allowing AI to diagnose UI errors and verify layouts.

Enables agents to select the provably optimal set of candidate actions under a budget, with optional conflicts and prerequisites, and to re-plan to the single best next action after each outcome. Also validates any proposed selection for feasibility.

Publishes a spending-summary tool alongside an interactive HTML view, letting people click a category in the answer to fetch the individual expense records behind that total. Keeps calculations and their supporting evidence together in a single inspectable flow.

Converts AI Skills (following Claude Skills format) into MCP server resources, enabling LLM applications to discover, access, and utilize self-contained skill directories through the Model Context Protocol. Provides tools to list available skills, retrieve skill details and content, and read supporting files with security protections.

A read-only MCP server for Anaplan that exposes 5 tools for business users to explore data through an AI assistant, with session caching and metadata gating.

A stdio MCP server that provides a single tool to query Gemini with Google Search grounding for current web information, returning synthesized answers with source links.

Enables Oracle Database schema introspection and SQL execution over stdio, letting MCP hosts list tables, describe columns, and run SQL—read-only by default with opt-in writes.

An MCP server that exposes the Wunder Mobility v2 end-user API as tools, enabling clients to browse vehicles, manage rentals, and create reservations on behalf of a signed-in user.

Enables MCP clients to send raw text prompts to inexpensive LLM models via the Devin/Windsurf backend, returning answers in result files to avoid context bloat, with support for multi-turn sessions and model discovery.

A cross-AI-agent local task context sharing MCP server that allows clients like Codex, Claude Code, Hermes, and VS Code to read and write the same task state, enabling task tracking and context distillation.