Skip to content
EngineeringOpenAlt AI Labs

An MCP server that gives Claude useful data about Kyrgyzstan

Claude path, stage 4. Build a Model Context Protocol server that lets Claude look up a real, public Kyrgyzstan dataset, and show Claude using it to answer a question.

  1. 1Learn
  2. 2Do
  3. 3Submit
  4. 4Results
Category
Engineering
Lesson
Self-paced
Effort
~8h
Prize
Free
Closes in
 

Learn

Take the course first

This stage starts with Introduction to Model Context Protocol, a free course by Anthropic on the Anthropic Academy site. It builds MCP servers and clients in Python. The course is in English and self-paced.

Anthropic Academy is made and run by Anthropic. Alt AI Labs is not affiliated with Anthropic: we link to the course and set this challenge.

Tools, resources and prompts

  • Tools are functions Claude can call, such as get_rate(currency, date).
  • Resources are data Claude can read, such as a list of holidays.
  • Prompts are ready-made instructions a user can pick.

This challenge needs at least two tools. Resources and prompts are welcome.

Descriptions do the work

Claude decides which tool to call from its name and description. "Returns the official National Bank exchange rate for a currency on a date, in som" works. "Gets data" does not.

Keep the data honest

Use a public source and say which one. If you typed the data up yourself, say where it came from and when.

Finish the lesson to unlock submitting.

Sign in to start learning

Do

about 8h of work

The task

The Model Context Protocol (MCP) is how you give Claude new tools and data without changing Claude. In this challenge you build an MCP server around one public dataset that matters here. Examples: official exchange rates, public holidays, a list of state services and the documents each needs, or bus routes you typed up yourself. The server exposes at least two tools that Claude can call, with clear descriptions and typed inputs. You connect it to Claude Desktop or Claude Code and record Claude using it to answer a real question.

Deliverables

  1. 01A public GitHub repo with the server code and a README with install and connect steps
  2. 02At least two tools, each with a description and typed input parameters
  3. 03Tests for each tool that run without Claude
  4. 04A screen recording of up to two minutes showing Claude calling your tools to answer a question

Acceptance criteria

Judged against exactly these

  • The server installs and connects to Claude Desktop or Claude Code by following the README
  • There are at least two tools, and each tool's description tells Claude when to use it
  • Each tool returns a clear error message for bad input instead of crashing
  • The data source is public, the README names it, and it is used within its terms
  • Tests run with one command and pass
  • The recording shows Claude choosing and calling the tools, not just the code running

Rules

  • Complete the lesson first. The submit form unlocks once you mark it done.
  • Submit a link to your work: a public GitHub repo or PR for code challenges; YouTube, Google Drive or Docs, Figma, Notion or a live site for everything else.
  • AI tools are allowed and encouraged. Plagiarised or copied work is disqualified.
  • The work must be made during the challenge window. For code, the commit history must show it.
  • An AI reviewer pre-scores every entry against the acceptance criteria. A human makes the final decision.
  • On payout, Alt AI Labs receives a licence to use the winning work (MIT for code). Non-winning entries stay yours.
  • Do not include secrets, API keys or anyone's personal data.
mcppythontypescriptapi-design

Submit

Sign in and complete the lesson to submit. Entry is free.

Sign in

Results

  1. 01

    AI pre-score

    Every entry is scored by an AI reviewer against the acceptance criteria.

  2. 02

    Human decision

    A person on the challenge team reviews the shortlist and makes the final call. The AI score is advisory.

  3. 03

    Win

    There is no cash prize. Winning work is shown to the organization behind the challenge.

Posted by

Alt AI Labs