ASKME-CLI

[!NOTE] We'd love your help! If you try askme-cli on Windows or have compatibility fixes, PRs are very welcome👏.

An MCP (Model Context Protocol) server that provides user confirmation interface for AI assistants

AI assistants often need user confirmation or next steps, but there's no simple way to get user input during MCP conversations.

This project provides a personal confirmation endpoint via terminal interface.

You will need to configure your MCP client to use ASKME-CLI.

Deploy the configuration using the instructions below.

# View help and configuration examples npx askme-cli help # Run directly (when called by MCP client) npx askme-cli

Add to your MCP client configuration file:

{ "mcpServers": { "askme-cli": { "command": "npx", "args": ["askme-cli"], "env": { "ASKME_CLI_TERMINAL": "warp" } } } }

{ "mcpServers": { "askme-cli": { "command": "npx", "args": ["askme-cli"], "env": { "ASKME_CLI_TERMINAL": "iterm2" } } } }

# Clone and install git clone <your-repo-url> cd askme-cli npm install # Build and run npm run build npm run start

Once configured, the AI assistant can use theask_metool to collect user confirmation:
- AI assistant callsask_metool with summary of completed work
- ASKME-CLI opens a new terminal window with confirmation interface
- User enters next plan or confirmation and submits
- AI assistant receives user input to continue

Configure viaASKME_CLI_TERMINALenvironment variable:

Collects user's next plan or confirmation.

- what_we_did(string): Summary of work completed by AI

- User's input for next plan or confirmation
- Supports text and image input

To use ASKME-CLI effectively in your AI assistant, add these rules to your system prompt:

- Must call the askme-cli.ask_me tool before ending any user prompt every time. - All prompts starting with /tk, don't modify code or files, and call the askme-cli.ask_me tool before ending. - When you have questions, always call the askme-cli.ask_me tool before ending. - If you have ideas for new features or changes, explain them clearly, analyze their feasibility, and ask if I want to proceed.

{ "tool": "ask_me", "parameters": { "what_we_did": "I've completed setting up the database schema and created the user authentication endpoints. The API is now ready for testing." } }

- TypeScript- Type-safe development
- React + Ink- Terminal UI framework
- MCP SDK- Model Context Protocol support
- Node.js- Runtime environment

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

Durable MCP control plane for long-running Codex Desktop tasks.

7 tools for AI agent cron job scheduling with SQLite persistence, retry logic, and structured observability. MIT license.

A development management tool for project planning, task management, and development workflows.

Multi-agent coordination over MCP: atomic gap-free claims, file leases, a shared ledger, presence, handoffs, and a task graph. Remote Streamable HTTP; self-host (AGPL) or hosted.

Exposes Makefile targets as callable tools for AI assistants.

A server for managing sequential development tasks with configurable rules using external .mdc files.

Project management inside every AI coding tool — plans what's next, tracks every decision, stops drift.

Facilitates software development planning through an interactive and structured approach.

A tool for structured software development planning, helping to break down projects into tasks and track progress.

The MCP server for Bitrix24 provides AI assistants with structured access to the Bitrix24 API. It delivers up-to-date method descriptions, parameters, and valid values, allowing assistants to work with precise data instead of guesswork. This reduces code errors and accelerates Bitrix24 integration development.