What are the requirements for FlowStepMCP?

FlowStepMCP requires the following environment variables: FLOWSTEP_MODE (optional), FLOWSTEP_TELEGRAM__BOTTOKEN (optional), FLOWSTEP_TELEGRAM__CHATID (optional). You'll also need a compatible MCP client like Claude Desktop or Claude Code.

Is FlowStepMCP free to use?

Yes, FlowStepMCP is open source and free to use. You can find the source code on GitHub.

What MCP clients support FlowStepMCP?

FlowStepMCP works with any MCP-compatible client including Claude Desktop (Anthropic's official desktop app), Claude Code (CLI tool), Cursor, and other editors with MCP support.

How do I configure FlowStepMCP?

Configure FlowStepMCP by adding it to your MCP client's config file. For Claude Desktop, edit claude_desktop_config.json and add the server configuration. See the Configuration section above for a ready-to-use example.

FlowStepMCP Server

Q: What tools does FlowStepMCP provide?

notify: Display non-blocking or blocking informational messages to the user.. confirm: Request explicit Yes/No or Cancel confirmation from the user.. select: Provide dropdowns or lists for choosing one or multiple options.. textInput: Collect free-form text from the user with multi-line support..

Q: How do I install FlowStepMCP?

Install FlowStepMCP by running: dotnet run -- --mode cli

Local setup required. This server has to be cloned and prepared on your machine before you register it in Claude Code.

Set the server up locally

Run this once to clone and prepare the server before adding it to Claude Code.

Run in terminal

git clone https://github.com/rodrigoramosrs/FlowStepMCP
cd FlowStepMCP

Then follow the repository README for any remaining dependency or build steps before continuing.

Register it in Claude Code

After the local setup is done, run this command to point Claude Code at the built server.

Run in terminal

claude mcp add flow-step -- node "<FULL_PATH_TO_FLOWSTEPMCP>/dist/index.js"

Replace <FULL_PATH_TO_FLOWSTEPMCP>/dist/index.js with the actual folder you prepared in step 1.

README.md

MCP server for seamless interaction between LLMs and end-users

FlowStep - MCP Server for User Interactions

🚧 Under Development

A complete Model Context Protocol (MCP) server designed to facilitate seamless interaction between Large Language Models (LLMs) and end-users. It provides a robust set of tools for notifications, confirmations, selections, and text inputs, supporting multiple rendering modes including Console, GUI, and Telegram Bot.

🎯 Native Desktop UI

▶️ Watch the demonstration video

FlowStep Demonstration

🎯 Telegram UI

▶️ Watch the demonstration video

🎯 Overview

FlowStep acts as an abstraction layer for user interactions. It exposes standard MCP tools that LLMs can invoke to interact with the user based on the application's configuration (Console, GUI, or Telegram).

Key Capabilities:

Notifications: Display non-blocking or blocking informational messages.
Confirmations: Request explicit Yes/No or Cancel confirmation from the user.
Single & Multi-Selection: Provide dropdowns or lists for choosing one or multiple options.
Text Input: Collect free-form text from the user with multi-line support.
Custom Input: Allow selection from a predefined list or custom text entry.
Progress Reporting: Visual feedback for long-running operations.
Multiple Render Modes: Console (CLI), Desktop GUI (Avalonia), or Telegram Bot.

📦 Project Structure

The library is organized into logical layers:

FlowStep.MCP.Library/
├── Models/
│   └── InteractionModels.cs           # Data models (InteractionRequest, InteractionResponse, InteractionOption)
├── Contracts/
│   ├── IFlowStepService.cs            # Core service interface
│   └── IInteractionRenderer.cs        # Renderer interface (Contracts for UI implementation)
├── Services/
│   └── FlowStepService.cs             # Business logic and orchestration
├── McpServices/
│   └── FlowStepMcpService.cs          # Implementation of MCP Server Tools
├── Renderers/
│   ├── CliInteractionRenderer.cs      # Console-based implementation
│   ├── TelegramRenderer.cs            # Telegram Bot implementation
│   ├── GuiInteractionBridge.cs        # Bridge for custom GUI implementations
│   └── AvaloniaUI/
│       ├── AvaloniaUIRenderer.cs      # Main Avalonia GUI renderer
│       ├── Themes/
│       │   └── ThemeColors.cs         # Dark mode color definitions
│       ├── Header/
│       │   └── HeaderContentFactory.cs
│       ├── Footer/
│       │   ├── StandardFooterFactory.cs
│       │   └── NotificationFooterFactory.cs
│       ├── Inputs/
│       │   ├── SingleChoiceInputFactory.cs
│       │   ├── MultiChoiceInputFactory.cs
│       │   ├── TextInputFactory.cs
│       │   └── ChoiceWithTextInputFactory.cs
│       ├── Factories/
│       │   ├── ConfirmationButtonsFactory.cs
│       │   ├── SimpleConfirmationContentFactory.cs
│       │   └── ResponseBuilder.cs
│       └── Styles/
│           └── DarkThemeStyles.cs     # XAML-like styling logic
├── Extensions/
│   └── FlowStepServiceExtension.cs    # DI Registration helper
└── FlowStep.MCP.Library.csproj

🖥️ Rendering Modes

FlowStep supports three rendering modes, configurable at startup:

Mode	Description	Use Case
CLI	Console/Terminal interface	Headless servers, debugging, automation scripts
GUI	Avalonia Desktop application	Rich desktop experience with modern dark UI
Telegram	Telegram Bot integration	Remote interactions, mobile notifications, distributed teams

Mode Selection Priority

Configuration is resolved in the following order (highest to lowest priority):

Command Line Arguments
Environment Variables (prefix: FLOWSTEP_)
appsettings.json
Default (GUI)

⚙️ Configuration

Command Line Arguments

# GUI Mode (default)
dotnet run

# CLI Mode
dotnet run -- --mode cli

# Telegram Mode
dotnet run -- --mode telegram --telegram-token "123456:ABC-DEF" --telegram-chat-id 123456789

# Custom configuration file
dotnet run -- --config /path/to/custom-config.json

Environment Variables

# Windows
set FLOWSTEP_MODE=telegram
set FLOWSTEP_TELEGRAM__BOTTOKEN=123456:ABC-DEF
set FLOWSTEP_TELEGRAM__CHATID=123456789

# Linux/Mac
export FLOWSTEP_MODE=telegram
export FLOWSTEP_TELEGRAM__BOTTOKEN=123456:ABC-DEF
export FLOWSTEP_TELEGRAM__CHATID=123456789

appsettings.json

{
  "Logging": {
    "LogLevel": {
      "Default": "Information",
      "Microsoft.AspNetCore": "Warning"
    }
  },
  "AllowedHosts": "*",
  "Mode": "gui",
  "Telegram": {
    "BotToken": "123456:ABC-DEF",
    "ChatId": "123456789"
  }
}

Environment-Specific Configuration

Create `appsetti

Tools (4)

notifyDisplay non-blocking or blocking informational messages to the user.

confirmRequest explicit Yes/No or Cancel confirmation from the user.

selectProvide dropdowns or lists for choosing one or multiple options.

textInputCollect free-form text from the user with multi-line support.

Environment Variables

FLOWSTEP_MODESets the rendering mode (cli, gui, or telegram)

FLOWSTEP_TELEGRAM__BOTTOKENTelegram bot token for Telegram mode

FLOWSTEP_TELEGRAM__CHATIDTelegram chat ID for notifications

Configuration

claude_desktop_config.json

{"mcpServers": {"flowstep": {"command": "dotnet", "args": ["run", "--project", "/path/to/FlowStep.MCP.Library.csproj"]}}}

Try it

→Ask the user for confirmation before deleting the generated file.

→Show a selection menu to the user to choose which database migration to run.

→Send a notification to my Telegram bot once the long-running data analysis is complete.

→Prompt the user for a custom input string to use as the project name.

Frequently Asked Questions

What are the key features of FlowStepMCP?

Supports multiple rendering modes: Console, Avalonia Desktop GUI, and Telegram Bot. Provides tools for notifications, confirmations, and multi-selection lists. Enables free-form text input collection from users. Allows custom input selection from predefined lists. Includes progress reporting for long-running operations.

What can I use FlowStepMCP for?

Headless server automation requiring occasional human approval. Desktop-based LLM workflows needing a rich UI for user interaction. Remote task management via Telegram bot notifications and inputs. Interactive CLI scripts that require user selection or confirmation.

How do I install FlowStepMCP?

Install FlowStepMCP by running: dotnet run -- --mode cli

What MCP clients work with FlowStepMCP?

FlowStepMCP works with any MCP-compatible client including Claude Desktop, Claude Code, Cursor, and other editors with MCP support.

Turn this server into reusable context

Keep FlowStepMCP docs, env vars, and workflow notes in Conare so your agent carries them across sessions.

Need the old visual installer? Open Conare IDE.

Open Conare

View all communication servers →

FlowStepMCP Server

Set the server up locally

Register it in Claude Code

FlowStep - MCP Server for User Interactions

🚧 **Under Development**

🎯 Native Desktop UI

🎯 Telegram UI

🎯 Overview

📦 Project Structure

🖥️ Rendering Modes

Mode Selection Priority

⚙️ Configuration

Command Line Arguments

Environment Variables

appsettings.json

Environment-Specific Configuration

Tools (4)

Environment Variables

Configuration

Try it

Frequently Asked Questions

Turn this server into reusable context

Related MCP Servers

🚧 Under Development