Telegram Bot MCP
Exposes Telegram Bot API operations as MCP tools, enabling managing messages, chats, and forum topics via a FastMCP server.
README
Telegram Bot MCP
A FastMCP server that exposes Telegram Bot API operations as Model Context Protocol (MCP) tools. The server uses aiogram for Telegram Bot API calls and pydantic-settings for environment-based configuration.
Requirements
- Python 3.10 or newer
- A Telegram bot token from BotFather
- An MCP-compatible client that can connect over
streamable-http,sse, orstdio
Installation
-
Clone or download this project.
-
Install dependencies.
pip install -r requirements.txt -
Create your environment file.
cp .env.example .envOn Windows PowerShell:
Copy-Item .env.example .env -
Edit
.envas needed for server transport, logging, and safety limits. (logging and safety limits currently not working)
Running the server
Run from the project root:
python telegram_bot_mcp.py
The default transport is streamable-http on 127.0.0.1:51000.
You can override transport, host, and port with command-line arguments:
python telegram_bot_mcp.py --transport streamable-http --host 127.0.0.1 --port 51000
Supported transport values are:
streamable-httpssestdio
MCP client configuration example
For a client that launches the server with a command, use a configuration similar to:
{
"mcpServers": {
"telegram-bot-mcp": {
"command": "python",
"args": ["telegram_bot_mcp.py", "--transport", "stdio"]
}
}
}
For HTTP-based clients, start the server with streamable-http and configure the client to connect to:
http://127.0.0.1:51000/mcp
Supported MCP tools
Server tools
| Tool | Description |
|---|---|
health_check |
Returns basic MCP server status, server name, and active transport. |
User and bot identity tools
| Tool | Parameters | Description |
|---|---|---|
get_me |
telegram_bot_token: string |
Returns sanitized identity information for the supplied Telegram bot token. |
Chat tools
| Tool | Parameters | Description |
|---|---|---|
get_chat_infomations |
telegram_bot_token: string, chat_id: int | string |
Returns sanitized Telegram chat information for a chat ID or username. Note: the tool name currently uses infomations as implemented in the project. |
delete_forum_topic |
telegram_bot_token: string, chat_id: int | string, message_thread_id: int, confirm: boolean = false |
Deletes a forum topic from a Telegram forum supergroup. Requires confirm=true. |
Message tools
| Tool | Parameters | Description |
|---|---|---|
delete_message |
telegram_bot_token: string, chat_id: int | string, message_id: int |
Deletes one Telegram message from a conversation. |
delete_messages |
telegram_bot_token: string, chat_id: int | string, message_ids: int[] |
Deletes multiple Telegram messages by explicit message IDs. Limited to 100 message IDs per call. |
delete_message_history_range |
telegram_bot_token: string, chat_id: int | string, start_message_id: int, end_message_id: int, confirm: boolean = false |
Best-effort deletion of every message ID in an inclusive range. Requires confirm=true and is limited to 100 message IDs per call. |
Telegram Bot API limitations
Telegram bots can only perform actions allowed by the Telegram Bot API and the bot's permissions in each chat. In particular:
- The bot must be a member of the target chat.
- Administrative permissions may be required for deleting messages or forum topics.
- The Telegram Bot API cannot enumerate an entire chat history for deletion.
- Message deletion is best-effort and may fail for messages the bot is not allowed to delete.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。