日本語版 READMEはこちら

LINE Bot MCP Server

Model Context Protocol (MCP) server implementation that integrates the LINE Messaging API to connect an AI Agent to the LINE Official Account.

Note

This repository is provided as a preview version. While we offer it for experimental purposes, please be aware that it may not include complete functionality or comprehensive support.

Tools

1. push_text_message
   • Push a simple text message to a user via LINE.
   • Inputs:
     • user_id (string?): The user ID to receive a message. Defaults to DESTINATION_USER_ID. Either user_id or DESTINATION_USER_ID must be set.
     • message.text (string): The plain text content to send to the user.
2. push_flex_message
   • Push a highly customizable flex message to a user via LINE.
   • Inputs:
     • user_id (string?): The user ID to receive a message. Defaults to DESTINATION_USER_ID. Either user_id or DESTINATION_USER_ID must be set.
     • message.altText (string): Alternative text shown when flex message cannot be displayed.
     • message.content (any): The content of the flex message. This is a JSON object that defines the layout and components of the message.
     • message.contents.type (enum): Type of the container. 'bubble' for single container, 'carousel' for multiple swipeable bubbles.
3. broadcast_text_message
   • Broadcast a simple text message via LINE to all users who have followed your LINE Official Account.
   • Inputs:
     • message.text (string): The plain text content to send to the users.
4. broadcast_flex_message
   • Broadcast a highly customizable flex message via LINE to all users who have added your LINE Official Account.
   • Inputs:
     • message.altText (string): Alternative text shown when flex message cannot be displayed.
     • message.content (any): The content of the flex message. This is a JSON object that defines the layout and components of the message.
     • message.contents.type (enum): Type of the container. 'bubble' for single container, 'carousel' for multiple swipeable bubbles.
5. get_profile
   • Get detailed profile information of a LINE user including display name, profile picture URL, status message and language.
   • Inputs:
     • user_id (string?): The ID of the user whose profile you want to retrieve. Defaults to DESTINATION_USER_ID.
6. get_message_quota

• Get the message quota and consumption of the LINE Official Account. This shows the monthly message limit and current usage.
• Inputs:
  • None

Installation (Using npx)

requirements:

• Node.js v20 or later

Step 1: Create LINE Official Account

This MCP server utilizes a LINE Official Account. If you do not have one, please create it by following this instructions.

If you have a LINE Official Account, enable the Messaging API for your LINE Official Account by following this instructions.

Step 2: Configure AI Agent

Please add the following configuration for an AI Agent like Claude Desktop or Cline.

Set the environment variables or arguments as follows:

• CHANNEL_ACCESS_TOKEN: (required) Channel Access Token. You can confirm this by following this instructions.
• DESTINATION_USER_ID: (optional) The default user ID of the recipient. If the Tool's input does not include user_id, DESTINATION_USER_ID is required. You can confirm this by following this instructions.

{
  "mcpServers": {
    "line-bot": {
      "command": "npx",
      "args": [
        "@line/line-bot-mcp-server"
      ],
      "env": {
        "CHANNEL_ACCESS_TOKEN" : "FILL_HERE",
        "DESTINATION_USER_ID" : "FILL_HERE"
      }
    }
  }
}

Installation (Using Docker)

Step 1: Create LINE Official Account

This MCP server utilizes a LINE Official Account. If you do not have one, please create it by following this instructions.

If you have a LINE Official Account, enable the Messaging API for your LINE Official Account by following this instructions.

Step 2: Build line-bot-mcp-server image

Clone this repository:

git clone [email protected]:line/line-bot-mcp-server.git

Build the Docker image:

docker build -t line/line-bot-mcp-server .

Step 3: Configure AI Agent

Please add the following configuration for an AI Agent like Claude Desktop or Cline.

Set the environment variables or arguments as follows:

• mcpServers.args: (required) The path to line-bot-mcp-server.
• CHANNEL_ACCESS_TOKEN: (required) Channel Access Token. You can confirm this by following this instructions.
• DESTINATION_USER_ID: (optional) The default user ID of the recipient. If the Tool's input does not include user_id, DESTINATION_USER_ID is required. You can confirm this by following this instructions.

{
  "mcpServers": {
    "line-bot": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "CHANNEL_ACCESS_TOKEN",
        "-e",
        "DESTINATION_USER_ID",
        "line/line-bot-mcp-server"
      ],
      "env": {
        "CHANNEL_ACCESS_TOKEN" : "FILL_HERE",
        "DESTINATION_USER_ID" : "FILL_HERE"
      }
    }
  }
}

Versioning

This project respects semantic versioning

See http://semver.org/

Contributing

Please check CONTRIBUTING before making a contribution.