Show HN: A “Course” as an MCP Server

#1

Post by codekarate » Tue, Jun 10, 2025, 8:36 PM UTC

We wanted to build a course for new Mastra devs to get started quickly. However, we knew videos would go out of date and be more difficult to maintain.

We decided to launch our "course" as an MCP server. This way your coding agent actually teaches the course content to you and can help you write the code. We think this is a really interactive way to learn.

Using an editor with MCP support (such as Cursor, Windsurf, or VSCode), your code agent will call the appropriate MCP tools which will return context for the agent. This context tries to instruct the agent that it should be teaching you the content, not just doing the work for you.

The course is still pretty experimental and some models work better than others. Code is available in the Mastra Github repo in the mcp-docs-server package (https://github.com/mastra-ai/mastra/tree/main/packages/mcp-d...)

Show HN: A “Course” as an MCP Server
mastra.ai

Re: Show HN: A “Course” as an MCP Server

#2

Post by avipeltz » Tue, Jun 10, 2025, 10:23 PM UTC

yall have been cooking ;) does the course also go through making a workflow or just an agent?

Re: Show HN: A “Course” as an MCP Server

#3

Post by codekarate » Tue, Jun 10, 2025, 10:41 PM UTC

avipeltz wrote: post #2

yall have been cooking ;) does the course also go through making a workflow or just an agent?

it currently only covers creating an agent, adding tools/mcp, and agent memory

later this week we will be releasing a workflows lesson

Re: Show HN: A “Course” as an MCP Server

#4

Post by hofo » Tue, Jun 10, 2025, 11:16 PM UTC

Newbie here: how much of creating agents are portable from one AI platform to another? Do they mostly have the same concepts but differ slightly in implementation? Or something else?

Re: Show HN: A “Course” as an MCP Server

#5

Post by avipeltz » Tue, Jun 10, 2025, 11:27 PM UTC

codekarate wrote: post #3

avipeltz wrote: post #2

yall have been cooking ;) does the course also go through making a workflow or just an agent?

it currently only covers creating an agent, adding tools/mcp, and agent memory later this week we will be releasing a workflows lesson

nice looking forward to it :)

Re: Show HN: A “Course” as an MCP Server

#6

Post by codekarate » Tue, Jun 10, 2025, 11:50 PM UTC

hofo wrote: post #4

Newbie here: how much of creating agents are portable from one AI platform to another? Do they mostly have the same concepts but differ slightly in implementation? Or something else?

The concepts are pretty similar but this course is definitely pretty Mastra specific.

If you are looking for a more general overview on agent development, my co-founder wrote a short book (available free at https://mastra.ai/book)

Re: Show HN: A “Course” as an MCP Server

#7

Post by davecyen » Wed, Jun 11, 2025, 1:30 AM UTC

This is really cool. I used the OG Codecademy to learn how to code, this is kind of like the next-gen version of that.

Re: Show HN: A “Course” as an MCP Server

#8

Post by codekarate » Wed, Jun 11, 2025, 1:53 AM UTC

davecyen wrote: post #7

This is really cool. I used the OG Codecademy to learn how to code, this is kind of like the next-gen version of that.

Thanks!

I definitely remember spending time on Codecademy years ago

Re: Show HN: A “Course” as an MCP Server

#9

Post by ninkendo » Wed, Jun 11, 2025, 3:00 AM UTC

Seeing the title I had the opposite assumption of what it was referring to: I thought of the interface to getting homework, taking tests, submitting assignments, etc as an MCP server and watching an LLM fully pass a course from an online university autonomously. I’m not convinced this isn’t possible today.

Re: Show HN: A “Course” as an MCP Server

#10

Post by jsemrau » Wed, Jun 11, 2025, 6:39 AM UTC

hofo wrote: post #4

Newbie here: how much of creating agents are portable from one AI platform to another? Do they mostly have the same concepts but differ slightly in implementation? Or something else?

Most agent frameworks implement a ReAct (Reflect->Act) pattern in the form of a loop that "reasons" until a "final_answer" has been found. Most framworks also implement some form of session cache (scratchpad), tool use, and thought logging/tracing.

While most implementations are similar they are not easily transferable.