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 889C a preview version. While we offer it for experimental purposes, please be aware that it may not include complete functionality or comprehensive support.
- push_text_message
- Push a simple text message to user via LINE.
- Inputs:
user_id(string): The user ID to receive a message. Defaults to DESTINATION_USER_ID.message.text(string): The plain text content to send to the user.
- push_flex_message
- Push a highly customizable flex message to user via LINE. Supports both bubble (single container) and carousel (multiple swipeable bubbles) layouts.
- Inputs:
user_id(string): The user ID to receive a message. Defaults to DESTINATION_USER_ID.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.
- 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.
requirements:
- Node.js v20 or later
Clone this repository:
git clone git@github.com/line/line-bot-mcp-server.git
Install the necessary dependencies and build line-bot-mcp-server when using Node.js. This step is not required when using Docker:
cd line-bot-mcp-server && npm install && npm run build
This MCP server utilizes a LINE Official Account. If you do not have one, please create it by following this instructions.
To connect to the Messaging API, you need to have a channel access token. You can confirm this by following this instructions.
Additionally, you will need the user ID of the recipient user for messages. You can confirm this by following this instructions.
Please add the following configuration for an AI Agent like Claude Desktop or Cline.
Insert the channel access token and user ID you obtained earlier into CHANNEL_ACCESS_TOKEN and DESTINATION_USER_ID, respectively.
Additionally, update the path to line-bot-mcp-server in mcpServers.args.
{
"mcpServers": {
"line-bot": {
"command": "node",
"args": [
"PATH/TO/line-bot-mcp-server/dist/index.js"
],
"env": {
"CHANNEL_ACCESS_TOKEN" : "FILL_HERE",
"DESTINATION_USER_ID" : "FILL_HERE"
}
}
}
}Build the Docker image first:
docker build -t line/line-bot-mcp-server .
{
"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"
}
}
}
}