Nick Taylor

Posted on Jul 29 • Edited on Aug 4 • Originally published at nickyt.co

Introducing the dev.to MCP server

#mcp #agenticai #devto #ai

If you've been wondering how to get your AI tools talking to dev.to's content without building yet another custom integration, I've got something that might interest you. I recently built a remote Model Context Protocol (MCP) server for dev.to that makes this whole thing a lot more straightforward.

nickytonline / dev-to-mcp

Dev.to MCP Server

A remote Model Context Protocol (MCP) server for interacting with the dev.to public API without requiring authentication.

Features

This MCP server provides access to the following dev.to public API endpoints:

get_articles - Get articles from dev.to with optional filters (username, tag, state, pagination)
get_article - Get a specific article by ID or path
get_user - Get user information by ID or username
get_tags - Get popular tags from dev.to
get_comments - Get comments for a specific article
search_articles - Search articles using query parameters

Installation

npm install
npm run build

Usage

The server runs as a remote HTTP server on port 3000 (or the PORT environment variable) and can be used with any MCP-compatible client.

npm start

The server will be available at http://localhost:3000 for MCP connections.

Development

# Build the project
npm run build
# Watch mode for development
npm run dev

# Linting
npm run

…

View on GitHub

Why? Well, a couple of reasons. One, I used to work at Forem, so dev.to is near and dear to my heart. The other reason is because I've been doing a lot of work in the MCP space recently.

pomerium / mcp-app-demo

Demo application showcasing how to build and secure MCP servers and clients with Pomerium using contextual access policies.

Welcome to the Pomerium Chat, a minimal chat application for showcasing remote Model Context Protocol servers secured with Pomerium.

quickstart.mp4

Pre-requisites

Linux or MacOS host
Docker and Docker Compose
Your machine should have port 443 exposed to the internet so that it could acquire TLS certificates from LetsEncrypt and OpenAI could call your MCP server endpoints.
OpenAI API Key

Quickstart

Environment Variables

Create a .env-mcp-app-demo file in the root directory and add the following environment variables:

OPENAI_API_KEY=your_api_key_here

Pomerium Config

Update pomerium-config.yaml and replace YOUR-DOMAIN with the subdomain you control. Create A DNS records for relevant hosts (or *.YOUR-DOMAIN).

By default, the access policy limits access to users with emails in YOUR-DOMAIN. See policy language reference if you need to adjust it.

Docker Compose

See docker-compose.yaml file in this repo.

docker compose up -d

It includes a demo SQLite server that requires a demo database, for example https://github.com/jpwhite3/northwind-SQLite3…

View on GitHub

What's MCP All About?

Let me back up a bit. The Model Context Protocol (MCP) is Anthropic's open standard for connecting AI models to external data sources.

MCP is like having a standardized way for any AI system to talk to external resources without reinventing the wheel every time. It's been dubbed "the USB-C for AI."

What dev-to-mcp Does

The dev-to-mcp server exposes dev.to's public API through six focused tools. Here's what you get:

Content Discovery:

get_articles - Filter by author, tag, popularity, publication state
search_articles - Full-text search across all of dev.to's content
get_tags - Access trending tags and topics

For example, ask to get the latest React articles

Content Access:

get_article - Grab specific articles by ID or URL path
get_user - Pull author profiles and user info
get_comments - Read through article discussions and comment threads

Here I am asking for my articles:

In VS Code:

In the Pomerium MCP app demo.

In Goose:

Real-World Use Cases

Once your AI has direct access to dev.to's ecosystem, you can do things like:

Research the latest React patterns by pulling top articles tagged with "react"
Track specific authors' posting patterns and expertise areas
Analyze community discussions to understand developer sentiment on new technologies
Generate content recommendations based on trending topics and engagement

The possibilities get interesting when you start combining different searches and analyses.

Getting It Running

The setup is pretty straightforward:

npm install
npm run build  
npm start

That's it. Your MCP server is now running on http://localhost:3000 and ready to serve up dev.to data. Use it in your favourite MCP client, like VS Code, Claude, Goose, etc.

For development, there's watch mode:

npm run dev

And if you want to deploy it via Docker:

docker build -t dev-to-mcp .
docker run -p 3000:3000 dev-to-mcp

However you run it, remember that the URL when you register it in an MCP client like VS Code will be e.g. http://localhost:3000/mcp

Under the Hood

The MCP server leverages the MCP TypeScript SDK and Vite for a clean, modern development experience and leverages the dev.to API for dev.to content.

modelcontextprotocol / typescript-sdk

The official TypeScript SDK for Model Context Protocol servers and clients

MCP TypeScript SDK

Overview

The Model Context Protocol allows applications to provide context for LLMs in a standardized way, separating the concerns of providing context from the actual LLM interaction. This TypeScript SDK implements the full MCP specification, making it easy to:

Build MCP clients that can connect to any MCP server
Create MCP servers that expose resources, prompts and tools
Use standard transports like stdio and Streamable HTTP
Handle all MCP protocol messages and lifecycle events

Installation

npm install @modelcontextprotocol/sdk

⚠️ MCP requires Node.js v18.x or higher to work fine.

Quick Start

Let's create a simple MCP server that exposes…

View on GitHub

What's Next

This is meant to be for the dev.to public API server. You have access to additional endpoints with a dev.to API key as well, but for the initial release, I omitted that to keep things simple.

It'd be neat if you could register an OAuth app to avoid API keys altogether. That way, it could be deployed with the additional user-specific endpoints without requiring an API key which is only works for the editable content of the owner of the API key if accessing the additional endpoints.

@ben, if you want to host it officially, I think that'd be awesome, and I'd also be happy to move it to the forem org if there's interest.

For those interested in trying it out if you don't feel like deploying it yourself, it's available at https://devto.mcp.maisonlab.dev. If you try to use it, you'll be denied, so if you're OK with sending me your email address, I can add you to the Pomerium policy to give you access.

That's it! Check out the repo and give it a star if you find it useful. 😎

If you want to stay in touch, all my socials are on nickyt.online.

Photo by Gerard Siderius on Unsplash

Top comments (27)

Ben Halpern • Jul 29

Very interesting stuff!

@ben, if you want to host it officially, I think that'd be awesome, and I'd also be happy to move it to the forem org if there's interest.

Will give a deep dive asap and give you an answer

Nick Taylor • Jul 30

Fayaz • Jul 30

Cool!

I too recently experimented with an mcp server implementation with dev.to API.
But I only did it with contest submissions!

Looks like you went full on with it. Super!

Use Algolia MCP Server & VSCode GitHub Copilot to Find and Submit Unique Projects in Dev Challenges! ❤️‍🔥

Fayaz ・ Jul 28

#devchallenge #algoliachallenge #webdev #ai

Nick Taylor • Jul 30

Noice!

Aditya • Jul 30 • Edited

Hi, your dev MCP server is awesome. I have also made MCP tools for dev.to. check it out- github.com/extinctsion/mcp-py-devto

And yes it would be awesome if we can Collab and make official dev.to MCP server under forem. @ben

Nick Taylor • Jul 30

Glad you like it!

Comment deleted

Nick Taylor • Jul 29

Initializing the sessions varies depending on your MCP client, e.g. VS Code, Goose, Claude etc.

not sure what you mean by sub requests. The MCP Client with the LLM will decide if it needs to call the tools in the dev.to MCP server.

Feel free to open an issue.

For example you can ask for the latest react articles. Notice how with the natural language, it knows to call the get_articles tool with the tag "react"