M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
L

LINE Bot MCP Server

Official
MCP server that connects an AI Agent to your LINE Official Account
How to install ↓ Compare GitHub source repository ↗
Category
Collaboration #8 of 55
Stars
★ 789 Very popular
Transport
stdio (local process)
Runtime
Node.js 22+ · Docker
Credentials
API key / credential required
License
Apache-2.0
Last commit
Tools
12
61FMRS · C

LINE Bot MCP Server is maintained by LINE and integrates its Messaging API with MCP; it exposes 12 tools covering message push and broadcast, user profiles, message quota, rich menu management and follower ID retrieval, and the repository is explicitly marked as a preview version. It suits users who already have a LINE Official Account and want an AI Agent to handle messaging and account operations, but the Channel Access Token must be protected carefully and broadcast or rich menu operations directly affect the live account.

Strongest · Documentation 14/20 Weakest · Reliability 10/20

Reliability
10/20
Security and permissions
11/20
Maintenance
13/20
Documentation
14/20
Setup experience
13/20
Why each score
Reliability 10/20
The README enumerates 12 tools with their input schemas, and each tool's stated purpose matches its parameters with no internal contradictions. However, this is a purely static review with nothing executed, and the repository shows no visible CI workflows or committed test files, so it is impossible to verify that the server starts cleanly, completes the MCP initialization and tool-list handshake, or that declared tools match real behavior. Error handling and dependency version control (e.g., @line/bot-sdk and MCP SDK versions) are unsupported by evidence. Per the static calibration rule, reliability cannot exceed 12 without verifiable execution evidence, so it scores 10 — the happy path looks plausible but tests and edge-case coverage are thin.
Security and permissions 11/20
Credential handling is reasonably sound: CHANNEL_ACCESS_TOKEN and the optional DESTINATION_USER_ID are passed via environment variables, and the sample config uses a FILL_HERE placeholder with no real tokens or secrets, so no red line is crossed. The tool set (push/broadcast messages, profile and quota lookup, rich menu management) is commensurate with the normal capabilities of the LINE Messaging API. Deductions: delete_rich_menu and set/cancel default rich menu are side-effecting operations with no documented confirmation or approval step; broadcast reaches all followers with potentially wide blast radius and no stated guard; create_rich_menu generates and uploads an image and auto-sets it as default, a heavy side effect with no confirmation. Data flows and external network boundaries are also undisclosed. No red line triggered, so 11.
Maintenance 13/20
License is Apache-2.0, which is clear, and the repo is not archived; 775 stars with only 9 open issues suggests some community attention and issue triage. The project declares conformance to semantic versioning and ships a CONTRIBUTING.md, indicating basic governance. But a static review cannot observe commit cadence, release history, dependency update policy, or a security response channel. The README self-describes the project as a preview version that may lack complete functionality and comprehensive support, implying limited maturity. Ownership is reasonably clear given the LINE organization. Score 13 — activity and governance are adequate but version/update path evidence is thin.
Documentation 14/20
Coverage is broad: two install paths (npx and Docker), required and optional environment variables, per-tool purpose and input parameters with optionality marked, external links for creating a LINE Official Account and enabling the Messaging API, and a local MCP Inspector debugging walkthrough. Sample configs are complete and steps are clearly numbered. Deductions: no disclosure of API quota costs or billing impact (a get_message_quota tool exists but limits are unexplained), no troubleshooting section, no statement of what the preview version lacks, and no warning about the blast radius of high-impact operations like broadcast. Usable overall but with hidden assumptions and troubleshooting gaps, so 14.
Setup experience 13/20
The install path is clear: only Node.js v22+ is needed, and a single-line npx config connects to Claude Desktop or Cline with a complete mcpServers JSON example; a Docker path with build and run commands is also given, and both mainstream client config examples are present. Environment variable setup is explicit and links to official guides for obtaining the token and user ID. Deductions: steps require the user to independently create the LINE account and enable the Messaging API on the LINE side, platform compatibility is undocumented, the Docker path requires cloning and building locally (manual steps), and no committed tests or CI evidence validate that the configuration actually works. Per the static calibration rule, setup cannot exceed 15 without execution evidence, so 13.

Static review · not runListed 2026-09-10

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkChanges third-party account data

Best for

  • Developers or operations teams that already have a LINE Official Account with the Messaging API enabled
  • Teams that want an AI Agent to automate LINE messaging and account operations
  • Users of MCP-capable clients such as Claude Desktop or Cline who need LINE integration

Not for

  • Users without a LINE Official Account or without the Messaging API enabled
  • Scenarios needing stable, fully supported production functionality beyond a preview release
  • Environments stuck below Node.js v22 that cannot use Docker

Required permissions

  • Requires CHANNEL_ACCESS_TOKEN (Channel Access Token) to call the LINE Messaging API
  • Optionally requires DESTINATION_USER_ID as the default recipient
  • Requires network access to the LINE Messaging API

Risks and side effects

  • CHANNEL_ACCESS_TOKEN is a sensitive credential; if leaked, others could send messages or read data on behalf of the Official Account
  • Broadcast messages reach all followers, so misuse can cause spam or an irreversible mass send
  • Operations such as create_rich_menu, delete_rich_menu and set_rich_menu_default directly change the Official Account's rich menu configuration
  • The repository is a preview version and may not include complete functionality or comprehensive support

Setup

Before you start

Runtime：Node.js 22+ · Docker

CHANNEL_ACCESS_TOKEN requiredsecret Long-lived Channel Access Token for your LINE channel, obtained from the LINE Developers Console; required.
Other optional settings (2)
DESTINATION_USER_ID optional Default recipient user ID, required when a tool input omits userId; follow LINE's docs to find your own user ID.
NPM_CONFIG_IGNORE_SCRIPTS optional npx setting to skip running package install scripts; not a credential.
  1. Create or prepare a LINE Official Account: if you do not have one, create it following the LINE instructions; if you already have one, enable the Messaging API for it.
  2. Using npx (requires Node.js v22 or later): add an mcpServers entry to your AI Agent such as Claude Desktop or Cline, with command "npx", args ["@line/line-bot-mcp-server"], and env containing NPM_CONFIG_IGNORE_SCRIPTS set to "true", CHANNEL_ACCESS_TOKEN (required) and DESTINATION_USER_ID (optional; required when the Tool's input does not include userId).
  3. Using Docker: clone the repository, build the image with docker build -t line/line-bot-mcp-server ., then configure command "docker" with args ["run", "-i", "--rm", "-e", "CHANNEL_ACCESS_TOKEN", "-e", "DESTINATION_USER_ID", "line/line-bot-mcp-server"] and put CHANNEL_ACCESS_TOKEN and DESTINATION_USER_ID in env.
  4. Local development with Inspector: clone the repository and cd into it, run npm install and npm run build, then run npx @modelcontextprotocol/inspector node dist/index.js -e CHANNEL_ACCESS_TOKEN="..." -e DESTINATION_USER_ID="..." to start the MCP Inspector for testing.
  5. Notes: CHANNEL_ACCESS_TOKEN can be confirmed via the LINE long-lived channel access token instructions; DESTINATION_USER_ID can be confirmed via the LINE get-user-IDs instructions.
claude_desktop_config.json 
{
  "mcpServers": {
    "line-bot": {
      "command": "npx",
      "args": [
        "@line/line-bot-mcp-server"
      ],
      "env": {
        "NPM_CONFIG_IGNORE_SCRIPTS": "true",
        "CHANNEL_ACCESS_TOKEN": "FILL_HERE",
        "DESTINATION_USER_ID": "FILL_HERE"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "line-bot": {
      "command": "npx",
      "args": [
        "@line/line-bot-mcp-server"
      ],
      "env": {
        "NPM_CONFIG_IGNORE_SCRIPTS": "true",
        "CHANNEL_ACCESS_TOKEN": "FILL_HERE",
        "DESTINATION_USER_ID": "FILL_HERE"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add line-bot -e NPM_CONFIG_IGNORE_SCRIPTS=true -e CHANNEL_ACCESS_TOKEN=FILL_HERE -e DESTINATION_USER_ID=FILL_HERE -- npx @line/line-bot-mcp-server

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After starting the client, the tool list should show the 12 recorded tools such as push_text_message and get_message_quota; calling get_message_quota and getting quota data confirms the connection works.

Troubleshooting

  1. Confirm Node.js is v22 or later
  2. Confirm CHANNEL_ACCESS_TOKEN is set correctly; if the tool input omits userId, also set DESTINATION_USER_ID
  3. When using npx, confirm NPM_CONFIG_IGNORE_SCRIPTS is set to "true" in env
  4. When using Docker, confirm the image was built with docker build and the environment variables are passed to the container correctly
  5. For local debugging, run npm run build first, then load dist/index.js with the MCP Inspector to verify the tools are available

Things to try

Once connected, you can ask your AI assistant things like:

  • Send a text message via LINE saying the meeting is moved to 3pm tomorrow
  • Check my LINE Official Account's monthly message quota and usage
  • List all rich menus on my account and set one as the default
  • Get the list of follower user IDs for my LINE Official Account

Tools 12

push_text_message writes
Push a simple text message to a user via LINE.
push_flex_message writes
Push a highly customizable flex message to a user via LINE.
broadcast_text_message writes
Broadcast a simple text message via LINE to all users who have followed your LINE Official Account.
broadcast_flex_message writes
Broadcast a highly customizable flex message via LINE to all users who have added your LINE Official Account.
get_profile read-only
Get detailed profile information of a LINE user including display name, profile picture URL, status message and language.
get_message_quota read-only
Get the message quota and consumption of the LINE Official Account, showing the monthly message limit and current usage.
get_rich_menu_list read-only
Get the list of rich menus associated with your LINE Official Account.
delete_rich_menu destructive
Delete a rich menu from your LINE Official Account.
Show 4 more tools
set_rich_menu_default writes
Set a rich menu as the default rich menu.
cancel_rich_menu_default writes
Cancel the default rich menu.
create_rich_menu writes
Create a rich menu based on the given actions, generate and upload an image, and set it as default.
get_follower_ids read-only
Get a list of user IDs of users who have added the LINE Official Account as a friend, allowing you to obtain user IDs for sending messages without manually preparing them.

Use cases

Let an AI Agent push text or flex messages to a specific user on behalf of the Official Account
Broadcast notices, announcements or campaigns to all followers
Retrieve user profiles to personalize replies
Check the monthly message quota and current usage of the Official Account
Create, set as default, or delete rich menus to manage the chat interface
Fetch follower user IDs in bulk for later messaging

Supported clients

Claude Desktop
Cline

Listed from the project's documentation, not tested by this site.

Overview

LINE Bot MCP Server is an official LINE implementation of a Model Context Protocol server that integrates the LINE Messaging API to connect an AI Agent to a LINE Official Account. It exposes capabilities for pushing and broadcasting messages, retrieving user profiles, checking message quota, managing rich menus, and fetching follower user IDs. The repository is marked as a preview version offered for experimental purposes, and it may not include complete functionality or comprehensive support.

Similar servers

Compare with other MCP servers

Source revision b17629e55bd6 Data synced 2026-10-19 Read the FMRS scoring method