Live Nov 18 · Cohort 3Nov 18, 19:00 MSK — free hands-on AI agents live workshop · Live Cohort 3 starts Nov 23Join Nov 18 Live →Cohort 3 (Nov 23)
ReleaseInsight · On RadarHabr AI→ original

How I Built 4 MCP Servers and a Protocol to Connect Them

A developer built four MCP servers for Claude instead of using off-the-shelf ones. The AI now works with Telegram on its own, searches for images on Civitai and manages advertising. To link all these services, he had to invent his own protocol.

AI-processed from Habr AI; edited by Hamidun News
How I Built 4 MCP Servers and a Protocol to Connect Them
Source: Habr AI. Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceHabr AI ↗
Editor-in-Chief
Key metric #1There was no Telegram integration at the scope I needed
Key metric #2There was no way to search content on Civitai (a prompt platform)
Key metric #3Managing ads in Yandex.Direct required a separate solution

I wrote four MCP servers, and a week later I realized they couldn't see each other. I had to invent an entire protocol to tie them together and build an automation pipeline.

What MCP is and why you need it

MCP (Model Context Protocol) is a protocol that lets AI assistants such as Claude connect to external services and work with them directly. In essence, it gives neural networks "hands."

Previously you had to write intermediate scripts, wrappers and integrations. Now it is enough to describe what your service can do, and Claude will call its methods itself, with no hacks and no intermediate layers.

Why the ready-made MCP servers didn't fit

Ready-made MCP servers exist for Slack, GitHub and Google Docs. But they weren't enough for my tasks.

  • There was no Telegram integration at the scope I needed
  • There was no way to search content on Civitai (a prompt platform)
  • Managing ads in Yandex.Direct required a separate solution
  • There was no tool for linking all the services into a single pipeline

It was simpler to write my own.

What I built

The result is four MCP servers that work together:

  • Telegram integration: reading posts, sending messages, managing channels
  • Civitai server: searching and downloading files with prompts
  • Yandex.Direct integration: managing campaigns and bids
  • Scheduler: planning future tasks and synchronization

Together they form a pipeline: Claude reads a task, finds the best content through Civitai, optimizes the campaign in Direct and publishes the result to three Telegram channels, all automatically.

"A programming classic: you start building one thing, wake up a week later and realize you've

written four servers and an entire architecture."

A custom protocol for communication

The problem: the four servers have to talk to each other. The Telegram server must tell the scheduler that a post has been published. The scheduler must tell Civitai what content to look for.

No ready-made standards existed for this kind of communication. I had to write my own protocol, a specification that defines how MCP servers see each other and exchange data. Now, when I add a new server, it plugs into the ecosystem automatically and starts interacting with the others.

What this means

MCP is no longer just a way to connect Claude to Slack or GitHub. It is a full architecture for building distributed automation systems in which AI coordinates the work of different services.

For developers, this means they can build complex pipelines without a central hub: Claude itself becomes the orchestrator.

Story

Key Questions Answered

What MCP is and why you need it?

MCP (Model Context Protocol) is a protocol that lets AI assistants such as Claude connect to external services and work with them directly. In essence, it gives neural networks "hands." Previously you had to write intermediate scripts, wrappers and integrations. Now it is enough to describe what your service can do, and Claude will call its methods itself, with no hacks and no intermediate layers.

Why the ready-made MCP servers didn't fit?

Ready-made MCP servers exist for Slack, GitHub and Google Docs. But they weren't enough for my tasks. There was no Telegram integration at the scope I needed There was no way to search content on Civitai (a prompt platform) Managing ads in Yandex.Direct required a separate solution There was no tool for linking all the services into a single pipeline It was simpler to write my own.

What I built?

The result is four MCP servers that work together: Telegram integration: reading posts, sending messages, managing channels Civitai server: searching and downloading files with prompts Yandex.Direct integration: managing campaigns and bids Scheduler: planning future tasks and synchronization Together they form a pipeline: Claude reads a task, finds the best content through Civitai, optimizes the campaign in Direct and publishes the result to three Telegram channels, all automatically.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Hands-On AI Engineering · Hamidun Academy

Build & Ship Production AI Agents Without Boilerplate Code

Start your free 72-hour trial at Hamidun Academy (academy.hamidun.com) — interactive lessons on Claude Code & Cursor, live Tech Radar, and production agent templates.

Start 72h Free Trial →Vibe-Coding Intensive
What do you think?
Loading comments…