Built an AI that uses block-code & how it works

Show and Tell
1

I built MCP Blockly for this hackathon, a full visual environment for creating real MCP servers with block based logic. Research shows that learners develop stronger understanding when they work hands on, so the goal here is to make MCP development something you can explore directly rather than only read about.

Under the hood, every block on the canvas is converted into live Python through a custom generator that rebuilds your MCP function signature, parameters, and logic on each edit. The AI assistant reads the entire workspace through a structured representation, plans multi step changes, creates and adjusts blocks, tests your tool with real inputs, and can even deploy the finished MCP server to your Hugging Face account.

Video:

Live Space:
<https://huggingface.co/spaces/MCP-1st-Birthday/MCP-Blockly>

If you like it, please leave a like on the space!

1 Like
2

Pretty cool project! Building an AI that uses block-based code seems like a great way to make coding more accessible. I’d love to know: which block-to-code system are you using, and can you share how the translation from blocks to executable code works under the hood?

1 Like
3

Thank you!

The project uses Blockly, an open source engine for making block-code systems. It has built in systems for turning blocks into code, and handling for things like statement blocks and value blocks. I’m not fully sure how it works for generating code, but I know how to use it.

If you’d like to know more about how other stuff (such as how the AI creates blocks and more) I’d love to explain!

cc @John6666

1 Like
4

Posted an article with the motivation for my project, along with things such as how it works!

1 Like