
Show HN:用 AI 控制你的 Telegram
這篇文章介紹了如何使用模型上下文協議(MCP)伺服器將 AI 與 Telegram 整合,讓使用者能以自然語言控制訊息、搜尋對話並草擬回覆,無需複雜的機器人設定。
Sign up
Sign in
Sign up
Sign in

Dev Genius
Follow publication

Coding, Tutorials, News, UX, UI and much more related to development
Follow publication
Control Your Telegram with AI: Complete Setup Guide for 2025

--
Listen
Share
Connect Telegram to AI via MCP — search, send, and control your messages with natural language
Your AI can now read, search, and send Telegram messages. Ask it to summarize a group chat, find all mentions of a project, or draft replies based on context. Set it up in 10 minutes.
What Is it ?
This is a Model Context Protocol (MCP) server that connects your entire Telegram account to AI assistants like Claude, Cline, Kiro CLI or any MCP-compatible client.
What you can do:
No bot creation. No complex APIs. Just your Telegram account + AI.
Why This Matters
Traditional Telegram bots are limited — they only work in specific chats and require complex setup. This gives you full access to your personal account through conversational AI.
Think of it as having a personal assistant who can read, search, and manage all your Telegram messages.
What You Need
See It In Action
Lets start with the fun part. Here’s what you can do:
List your chats:
“Show me my recent Telegram chats”
Search messages:
“Search recent news by New York times”
Send a message:
“Send a message to the Dev Team chat saying ‘Deployment complete ✅’”
Get chat summaries from multiple channels:
“Summarize last day crypto news from my telegram crypto related channels”
Real-World Use Cases
Here’s how people are using this:
Get Orrbenyamini’s stories in your inbox
Join Medium for free to get updates from this writer.
For Remote Teams:
For Crypto Traders:
For Developers:
For Productivity:
How to Set It Up
Step 1: Get Telegram API Credentials
First, you need API credentials from Telegram. Don’t worry — this is official and takes 2 minutes.
- Save your api_id and api_hash
Step 2: Install the Server
Clone and set up the server:
Create your environment file:
Edit .env with your credentials:
Important: Use your full phone number with country code (e.g., +1234567890).
Step 3: Authenticate with Telegram
Before using the server, authenticate with Telegram using the standalone auth script:
You’ll see a prompt asking for a verification code. Check your Telegram app — you’ll receive a code. Enter it.
If you have 2FA enabled, you’ll also need to enter your password.
That’s it! Your session is saved in the session/ directory and will be reused automatically by the MCP server. You won't need to authenticate again.Step
4: Connect to Your AI Client
Now connect the server to your AI assistant. Here’s how for the most popular clients:
For Claude Desktop
Edit your config file:
Add this:
For Amazon Q CLI
Edit ~/.config/mcp/mcp.json:
Restart your AI client and you’re ready!
Troubleshooting
“Invalid phone number”
“Authentication failed”
“MCP server not connecting”
“Can’t receive verification code”
Try It Yourself
⭐ Star the project on GitHub to stay updated
🤝 Contributions welcome — found a bug or have an idea? Open an issue!
💬 What would you automate? Drop your use case in the comments below.
Quick Links
Built with the Model Context Protocol — the open standard for connecting AI assistants to external tools and data sources.
Tags: #Telegram #AI #Automation #MCP #OpenSource #Claude #Productivity
--
--


Published in Dev Genius
Coding, Tutorials, News, UX, UI and much more related to development


Written by Orrbenyamini
No responses yet
Help
Status
About
Careers
Press
Blog
Privacy
Rules
Terms
Text to speech
相關文章