Build a Custom MCP Server for Cursor

Observe AI Agents built with Google’s ADK [Open-source]

Recently, Google released its new Agent Development Kit (ADK). Opik is the first observability platform to support it natively.

If you are building agents on ADK, you know visibility is everything.

With Comet’s Opik, you can now evaluate, monitor, and debug agents in real time—no extra setup required.

That said, Opik also integrates with other popular frameworks like CrewAI, LlamaIndex, and many more.

Check Opik’s documentation here →

Thanks to Comet for partnering today!

---

Build a Custom MCP Server for Cursor

Lately, there has been a lot of interest in MCPs.

Today, we are sharing a step-by-step video walkthrough (check above) of creating your own MCP server!

It connects to Cursor and lets it perform deep web searches using the LinkUp web search tool, as well as RAG over a specified directory.

Cursor IDE is our MCP host, and we connect it to our custom server.

Here’s the tech stack:

• Linkup for deep web search.

• Llama Index workflow to enable RAG.

​Find the code on GitHub here → (don't forget to star the repo)​

If you want to learn more about MCPs, here are some resources:

• Function calling & MCP for LLMs:

  

  • Before MCPs became popular, most AI workflows relied on traditional Function Calling.

  • Now, MCP (Model Context Protocol) is introducing a shift in how developers structure tool access and orchestration for Agents.

  • The visual above explains Function calling & MCP.

  • Read the full issue here →

• MCP-powered Agentic RAG:

  

  • In this build, we showcased a demo with MCP—an Agentic RAG.

  • It searches a vector database and falls back to web search if needed.

  • Read the full issue here →

• 5 powerful MCP servers for AI Agents →

  • Firecrawl MCP server

  • Browserbase MCP server

  • Opik MCP server

  • Brave MCP server

  • Sequential thinking MCP server

Thanks for reading!

---

P.S. For those wanting to develop “Industry ML” expertise:

At the end of the day, all businesses care about impact. That’s it!

• Can you reduce costs?

• Drive revenue?

• Can you scale ML models?

• Predict trends before they happen?

We have discussed several other topics (with implementations) that align with such topics.

Here are some of them:

• Learn how to build Agentic systems in an ongoing crash course with 11 parts.

• Learn how to build real-world RAG apps and evaluate and scale them in this crash course.

• Learn sophisticated graph architectures and how to train them on graph data in this crash course.

• So many real-world NLP systems rely on pairwise context scoring. Learn scalable approaches here.

• Learn how to run large models on small devices using ​Quantization techniques.

• Learn how to generate prediction intervals or sets with strong statistical guarantees for increasing trust using ​Conformal Predictions.

• Learn how to identify causal relationships and answer business questions using causal inference in this crash course.

• Learn how to scale and implement ML model training in this practical guide.

• Learn techniques to reliably test new models in production.

• Learn how to build privacy-first ML systems using ​Federated Learning.

• Learn 6 techniques with implementation to compress ML models.

All these resources will help you cultivate key skills that businesses and companies care about the most.