Telegram Reminder MCP
Enables users to manage Telegram reminders via natural language or commands, with features like setting, updating, deleting, and listing reminders, while ensuring multi-user privacy.
README
Telegram Reminder MCP
A context-aware, secure, and intuitive Telegram Reminder Bot powered by FastMCP and SQLite3. Set, view, modify, and delete reminders using simple conversational phrases or standard commands.
🤖 Live Bot & Web Preview
- Telegram Bot: Reminder_MCP_Bot
- Web Dashboard: reminder.minecraftbedrock.dpdns.org
✨ Features
- Natural Language Parsing: Recognizes relative phrases to set reminders (e.g.,
in 5 minutes,after 2 hours,in 3 days). - Full CRUD Capabilities: View, update, or cancel your reminders directly from Telegram chat messages.
- Strict Privacy & Multi-User Isolation: Reminders are strictly isolated using Telegram
chat_id. A user can never view, edit, or delete reminders belonging to another Telegram user. - Local Timezone Alignment: Configured specifically to calculate and store times in the Asia/Kolkata (IST) timezone.
- Robust Cron Engine: A background worker polls the database every 5 seconds to ensure alerts are dispatched down to the second.
- Voice Message Support & Normalization: Send voice notes directly in Telegram to set, list, edit, or delete reminders. Spoken number words (like "five" or "twenty-three") are automatically normalized to digits (like "5" or "23") to integrate seamlessly with the command parser.
💬 Sentence Tips & Parser Syntax
The bot supports natural conversational inputs. The syntax parser uses flexible regular expressions under the hood.
Setting a Reminder
You can set reminders using two main sentence structures:
remind me after [number] [unit] to [action]- Example:
remind me after 15 minutes to drink water - Example:
remind me after 1 hr to stretch
- Example:
remind me to [action] in [number] [unit]- Example:
remind me to call dad in 2 days - Example:
remind me to stretch in 30 seconds
- Example:
Supported Units: sec, second, seconds, min, minute, minutes, hr, hour, hours, day, days.
Voice Message Commands
You can tap the microphone icon in Telegram to speak any command. Spoken numbers are converted to digits automatically:
- Voice Input: "remind me in ten minutes to do yoga" -> bot normalizes to
"remind me in 10 minutes to do yoga"and schedules it. - Voice Input: "list reminders" -> bot transcribes and returns active reminders.
- Voice Input: "delete reminder four" -> bot normalizes to
"delete reminder 4"and deletes it. - Voice Input: "edit reminder five to walk dog in twenty minutes" -> bot normalizes to
"edit reminder 5 to walk dog in 20 minutes"and updates it.
Updating a Reminder via Chat
edit reminder [id] to [new text]oredit [id] to [new text]- Example:
edit reminder 11 to go to gym in 5 minutes - Example:
edit 11 to buy juice
- Example:
Deleting a Reminder via Chat
delete reminder [id],delete [id], orremove [id]- Example:
delete reminder 11
- Example:
Listing Reminders via Chat
list reminders,list,list all, orshow reminders
🛠️ Command Reference
For users who prefer classic bot commands:
| Command | Usage | Description |
|---|---|---|
/start |
/start |
Welcomes the user and lists available options. |
/help |
/help |
Displays the usage guidelines and tips. |
/list |
/list |
Displays a detailed list of all your active reminders and their remaining time. |
/delete |
/delete <id> |
Instantly cancels and removes the specified reminder from the database. |
/edit |
/edit <id> <new text and/or time> |
Modifies the task text or time offset for an active reminder. |
🔒 Security & Multi-User Privacy Check
All reminder actions (reading, updating, deleting) are secured on the FastMCP Server by verifying that the requesting user's Telegram chat_id matches the owner of the target reminder ID in the database.
- Isolation Test: If
User Aattempts to rundelete reminder 5or/delete 5, but the reminder belongs toUser B, the server returns anUnauthorizederror and blocks the transaction. - Result: Reminders are strictly confidential, and no cross-user visibility is possible.
🚀 Technical Setup (Local Deployment)
1. Requirements
- Python 3.9+
- FastMCP
- SQLite3
2. Configuration
Copy the .env.template to .env and fill in your details:
TELEGRAM_BOT_TOKEN=your_telegram_bot_token
MCP_SERVER_URL=http://127.0.0.1:9876/mcp
3. Run Server
Start the database server:
python3 server.py
4. Run Client
Start the Telegram Bot polling and background cron checker:
python3 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。