Own product · AI and MCP

MCPOrbital: an open source MCP server for space data

Language models tend to make up numbers about space. MCPOrbital is my own MCP server that connects Claude, Cursor and other AI assistants to real data from NASA JPL, NOAA, CelesTrak, the NASA Exoplanet Archive and arXiv, with a source on every answer.

Type
Own product, open source
For
Researchers and people building AI tools
Sources
NASA JPL, NOAA, CelesTrak, NASA Exoplanet Archive, arXiv
Delivered
MCP server, website, skills for AI agents
Period
May to June 2026
AI assistant answering about asteroid Apophis and space weather with NASA and NOAA sources
Real answers from MCPOrbital on 31 October 2026, shown in a simplified chat window.
In short

MCPOrbital is an open source MCP server with 11 tools: asteroids and close approaches, impact risk, space weather, satellite orbits, ephemerides and exoplanets. Ask an AI assistant when Apophis comes closest to Earth, and it fetches the answer from NASA JPL: 13 April 2029 at 21:46 UTC, about 38,011 km from Earth, with the source next to it.

11tools for AI assistants
42APIs researched before building
5open data sources, cited on every answer

Why

An AI assistant that answers from memory can get dates, distances and risks wrong. An MCP server gives it tools to fetch fresh numbers from the source instead of guessing, and to show where the numbers come from.

What I built

  1. Research first. 12 documents on APIs, market, the MCP spec, architecture and roadmap, and 42 APIs researched.
  2. An MCP server in TypeScript. 11 tools with input validation, and source, license and timestamp on every answer.
  3. Safety. The server refuses requests to command satellites.
  4. A website. Next.js with 12 pages and 13 API routes, where research briefs export as .md or .bib.
  5. Agent skills. A skill for Claude and one for OpenClaw with four agents.
  6. Hosting. On my own server with systemd, nginx and Let's Encrypt, with a public health check.

The result

The server is live with 11 tools, and the code is open on GitHub under the Apache 2.0 license. The answers in the illustration above are real answers from the server on 31 October 2026.

The same technique can connect an AI assistant to your CRM, your bookings or your prices, so it answers with your numbers instead of guessing.

Tools and tech stack

  • TypeScript
  • Node.js
  • MCP SDK
  • zod
  • Next.js
  • Tailwind CSS
  • nginx
  • Let's Encrypt

Want this for your business? See Generative engine optimization (GEO) or AI automation services.

Questions and answers

What is an MCP server?

MCP (Model Context Protocol) is an open standard that lets AI assistants like Claude use tools and data outside the model. An MCP server offers such tools, for example fetching data from an API.

Why cite a source on every answer?

So you can check the number yourself, and the AI assistant can show where the answer comes from. That makes the answers safer to use in research and work.

Can an MCP server connect AI to my business data?

Yes. The same approach can give an AI assistant access to your inventory, prices, open time slots or CRM customers, so it answers staff and customers with correct numbers.

Can I use MCPOrbital?

Yes, the code is open on GitHub under the Apache 2.0 license.

About the author

Vladyslav Lysechko
Vladyslav Lysechko, Performance marketer, Lysechko Digital, Rånåsfoss, Nes, Norway

I have worked with ads on Google, Meta and TikTok for more than seven years and personally managed ad budgets from $1,000 a day up to $50,000 a month. For the last two years I have built websites, automations and AI assistants with AI every day. I help local businesses in Oslo and on Romerike get more calls and bookings. LinkedIn

More case studies

AI and automationAds and tracking

Eva: AI marketing agent

An AI agent that reads the team chat, tracks cost per qualified lead and pauses, scales and launches ads within fixed limits.

64ad sets paused in two weeks

Read the case study

Systems and dashboardsAI and automation

Kanban task board for a software team

A Trello style board on its own server, with realtime updates for the whole team, a roadmap map and a docs library.

100tasks on the board in May

Read the case study

Systems and dashboardsAI and automation

Online course platform with CRM

A self hosted Teachable alternative: 282 lessons moved with zero errors, and access opens automatically after payment.

282lessons and 281 videos moved with zero errors

Read the case study

All case studies

Let's have a short call

Tell me a little about your business and you get a concrete suggestion with a fixed price.

Email vlad@lysechko.com · Phone +47 462 17 168

Book a short call WhatsApp (opens in a new tab)