VK to Telegram bridge MCP

vk-tg-bridge

A 4-tool MCP server that pulls VK wall posts, filters them by rules and publishes them to Telegram via the Bot API

  • 0 ★ on the repository
  • Last commit
  • MIT
MCP server

Broad access

Why this level

  • Publishes content to a Telegram channel on behalf of a bot
  • Reads and forwards VK wall content without human review when bridge_run is called end to end
All reasons and checks
Russian stack

shekelstrong/vk-tg-bridge

Install

Services

Manual install

git clone https://github.com/shekelstrong/vk-tg-bridge.git && cd vk-tg-bridge && pip install -r requirements.txt

Installs dependencies; VK_TOKEN and TG_BOT_TOKEN are needed afterwards.

Access and checks

How we review

Broad accessThe tool can change data, run commands or handle secrets.

  • Publishes content to a Telegram channel on behalf of a bot
  • Reads and forwards VK wall content without human review when bridge_run is called end to end

This is third-party code. Review the repository files before installing.

What it does

The server solves one job: moving content from VK to Telegram. fetch_vk pulls posts from a user's or community's wall via the VK API, filter_posts screens them by text length, include/exclude keywords, media presence and a minimum like count, format_tg prepares the text for Telegram with HTML markup and length trimming, and bridge_run chains all three steps and sends the filtered posts to a channel or chat through the Telegram Bot API. Each tool can be called on its own or as a single pipeline.

Who it is for. For people mirroring content from a VK group into a Telegram channel who want ads and weak posts filtered out automatically.

Good fit when

  • You want to automatically move the best posts from a VK group into a Telegram channel
  • You need keyword filtering, for example excluding posts containing the word ads
  • You want to test fetch and filter separately before publishing to Telegram

Not a fit when

  • You need the reverse direction, from Telegram into VK, the server only works one way
  • You need publishing with complex formatting, format_tg only does basic HTML and trimming

Example request

Bridge today's posts from group -123456 into the Telegram channel where likes are above 50 and there is no ad content

MCP

Transport
stdio
Authentication
API key

Environment variables

VK_TOKEN required, secret
A VK access token for reading a user's or community's wall.
TG_BOT_TOKEN required, secret
A Telegram bot token for sending posts to a channel or chat.

Limitations

It needs a separate VK_TOKEN and TG_BOT_TOKEN. The project is small, and the README does not describe duplicate handling on repeated bridge_run calls or VK API rate limits.

How to disable. Stop the server process and remove it from your MCP client configuration.

FAQ

Can I preview what gets filtered before sending?

Yes, fetch_vk and filter_posts can be called separately from bridge_run to inspect the intermediate result before publishing.

What filter rules are supported?

Minimum and maximum text length, include and exclude keywords, media presence, and a minimum like count.

Discussion

No questions or reviews yet.

Russian stack

An MCP server publishes one text at once to Telegram, VKontakte (group and personal wall) and TenChat with a single publish_all call

MCP serverBroad access
Russian stack

An MCP server for the VKontakte wall: reading and posting, messages, friends, communities and group stats in 8 tools

MCP serverBroad access1Repository stars
Russian stack

Remote MCP server for VK communities: 50 tools, posts, comments, dialogs, photos, video and products, with writing off by default

MCP serverBroad access
Russian stack

MCP server auto-generated from the VK API schema: over 180 tools with filters by read, write, money modes and API sections

MCP serverBroad access1Repository stars

README in short

The README describes the server as a VK-to-Telegram bridge with four tools, showing an example call and response shape for each: fetch_vk returns a post list, filter_posts reports how many passed versus were skipped, format_tg returns the ready text and parse mode, and bridge_run ties it together with counts of fetched, filtered and sent posts. The repository structure shows a separate file per tool and a CI workflow.

README badge

Report a problem