mcp-telegram and fast-mcp-telegram

The two tools are competitors, with tool A being a more feature-rich and deployment-ready successor or alternative to tool B, given its additional features like multi-user auth, intelligent search, file sending, web setup, and Docker/PyPI readiness.

mcp-telegram
62
Established
fast-mcp-telegram
59
Established
Maintenance 2/25
Adoption 17/25
Maturity 25/25
Community 18/25
Maintenance 10/25
Adoption 14/25
Maturity 18/25
Community 17/25
Stars: 237
Forks: 35
Downloads: 1,198
Commits (30d): 0
Language: Python
License: MIT
Stars: 29
Forks: 9
Downloads: 1,327
Commits (30d): 0
Language: Python
License: MIT
Stale 6m
No risk flags

About mcp-telegram

dryeab/mcp-telegram

MCP Server for Telegram

Implements Telegram's MTProto protocol via Telethon to expose messaging, search, draft management, and media download capabilities as MCP tools for AI agents. Integrates with Claude Desktop and Cursor through stdio transport, with credentials managed via interactive CLI login and session-based authentication. Provides LLMs with 10+ tools for sending/editing/deleting messages, searching dialogs, and downloading media while handling SQLite session locking for concurrent instance safety.

About fast-mcp-telegram

leshchenko1979/fast-mcp-telegram

Telegram MCP Server and HTTP-MTProto bridge | Multi-user auth, intelligent search, file sending, web setup | Docker & PyPI ready

Implements a dual-transport MCP server exposing Telegram's MTProto API with automatic entity resolution and safety guardrails, enabling both stdio-based development and HTTP-authenticated production deployments. The architecture consolidates functionality into fewer, context-efficient tools (e.g., unified `get_messages` for search/browse/read-by-ID/replies) following an LLM-optimized design philosophy, while the HTTP-MTProto bridge provides raw API access via curl for edge cases. Built on async Telethon with session persistence, LRU cache management, and structured logging for reliability.

Scores updated daily from GitHub, PyPI, and npm data. How scores work