Line

​

Required Environment Variables

• DESTINATION_USER_ID – (string) LINE user ID or group to receive messages.
• CHANNEL_ACCESS_TOKEN – (string) LINE Official Account channel access token.

​

Steps

1. Install Meshagent and dependencies:
   Copy

   Ask AI

   pip install "meshagent[all]"
   

2. Sign up and authenticate: Follow instructions at Meshagent CLI: Getting Started.
3. Start an MCP server agent in the test room:
   Copy

   Ask AI

   meshagent service test --room=test --role=agent --image=meshagent/mcp-line:latest \
     --env MESHAGENT_PORT=8001 \
     --env DESTINATION_USER_ID=FILL_HERE \
     --env CHANNEL_ACCESS_TOKEN=FILL_HERE \
     --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
     --name=mcp-line-service-test
   

   • This starts a Meshagent Room named test with a test MCP server running in it. Rooms auto-close if inactive.
4. Join a chatbot to the test room:
   Copy

   Ask AI

   meshagent chatbot join --room=test --agent-name=mcp-line --name=mcp-line --toolkit=mcp-line
   

   • Starts a chatbot that can use the LINE MCP server toolkit in the same room.
   • The command output provides a link to the room UI.
5. Test the setup:
   • Open the room link in your browser and send messages to the agent to interact with the MCP server tools.

​

Project-Level Deployment (for Persistent Production Use)

1. Create a persistent MCP server agent for the project:
   Copy

   Ask AI

   meshagent service create --role=agent --image=meshagent/mcp-line:latest \
     --env MESHAGENT_PORT=8001 \
     --env DESTINATION_USER_ID=FILL_HERE \
     --env CHANNEL_ACCESS_TOKEN=FILL_HERE \
     --port="num=8001 path=/webhook liveness=/ type=meshagent.callable" \
     --name=mcp-line-service
   

2. Create a persistent chatbot service:
   Copy

   Ask AI

   meshagent service create --image="meshagent/cli:latest" \
     --port="num=9001 path=/agent liveness=/ type=meshagent.callable participant_name=mcp-line-chatbot" \
     --env="MESHAGENT_PORT=9001" \
     --name="mcp-line-chatbot-service" \
     --command="meshagent chatbot service --agent-name=mcp-line-chatbot --toolkit=mcp-line"
   

​

Tools Available

​

Tool: broadcast_flex_message

• message (object)

​

Tool: broadcast_text_message

• message (object)

​

Tool: cancel_rich_menu_default

​

Tool: delete_rich_menu

• richMenuId (string): The ID of the rich menu to delete.

​

Tool: get_message_quota

​

Tool: get_profile

• userId (string, optional): The user ID to get a profile. Defaults to DESTINATION_USER_ID.

​

Tool: get_rich_menu_list

​

Tool: push_flex_message

• message (object)
• userId (string, optional): The user ID to receive the message. Defaults to DESTINATION_USER_ID.

​

Tool: push_text_message

• message (object)
• userId (string, optional): The user ID to receive the message. Defaults to DESTINATION_USER_ID.

​

Tool: set_rich_menu_default

• richMenuId (string): The ID of the rich menu to set as default.

​

Resources

• Meshagent: https://www.meshagent.com/
• Meshagent Documentation: https://docs.meshagent.com
• Meshagent MCP Tool Image: https://hub.docker.com/r/meshagent/mcp-line
• MCP Server (LINE) Repo: https://github.com/line/line-bot-mcp-server
• What is an MCP Server? https://www.anthropic.com/news/model-context-protocol
• Why use Docker for MCP Servers? https://www.docker.com/blog/the-model-context-protocol-simplifying-building-ai-apps-with-anthropic-claude-desktop-and-docker/