telegram-mcp
Enables interaction with Telegram Bot API for sending messages, photos, editing messages, answering callbacks, and fetching updates.
README
telegram-mcp
Dedalus MCP server for Telegram Bot API workflows. It exposes bot identity, message sending, photo sending, update polling, callback answers, and message editing.
Tools
get_me- Get bot identity.send_message- Send a text message to a chat.send_photo- Send a photo by URL or file ID.get_updates- Fetch bot updates.answer_callback_query- Respond to an inline keyboard callback query.edit_message_text- Edit a bot message.
Safety Notes
send_message, send_photo, answer_callback_query, and edit_message_text are user-visible. They can contact real Telegram users or groups and may trigger notifications. Confirm chat_id, message_id, callback IDs, and message contents before calling write tools.
Telegram bot tokens grant control of the bot. Do not commit tokens or chat exports.
Important DAuth Limitation
Telegram's official Bot API requires the token in the request URL:
https://api.telegram.org/bot<TOKEN>/<method>
dedalus_mcp 0.7.0 Connection supports secret injection into HTTP headers, not URL path templating. This repository still declares a DAuth connection named telegram-mcp with secret key TELEGRAM_BOT_TOKEN, but the implemented HTTP calls use a server-side TELEGRAM_BOT_TOKEN environment variable until path-secret support is available.
This means the current code is best deployed as a Type 2/3 server with a server or required credential environment value, not as a fully isolated path-templated DAuth server. Revisit this once Dedalus supports path credential templating for Telegram.
Environment
TELEGRAM_BOT_TOKEN=123456:ABC-DEF...
DEDALUS_AS_URL=https://as.dedaluslabs.ai
For _client.py:
DEDALUS_API_KEY=dsk-live-...
DEDALUS_API_URL=https://api.dedaluslabs.ai
DEDALUS_MCP_SERVER=dedalus-labs/telegram-mcp
TELEGRAM_TEST_CHAT_ID=...
TELEGRAM_TEST_MESSAGE_ID=...
TELEGRAM_TEST_CALLBACK_QUERY_ID=...
Run Locally
uv run python src/main.py
In another terminal:
MCP_SERVER_URL=http://127.0.0.1:8080/mcp uv run python src/_client.py
The client calls every tool. It intentionally includes write calls, so use a safe test chat.
Deploy
Publish as dedalus-labs/telegram-mcp, configure TELEGRAM_BOT_TOKEN securely, deploy, then run:
DEDALUS_MCP_SERVER_ONLY=1 DEDALUS_MCP_SERVER=dedalus-labs/telegram-mcp uv run python src/_client.py
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。