michaaels commented on issue #9052:
URL: https://github.com/apache/hop/issues/9052#issuecomment-6303357152

   Hi Matt,
   
   I've been working on a similar initiative over the past few months, so I was 
glad to see this issue come up.
   
   I've been developing an independent MCP connector for Apache Hop, with the 
longer-term goal of contributing the relevant functionality back to the main 
Apache Hop project once the implementation is mature and robust enough.
   
   Here's the repository:
   
   https://github.com/michaaels/hop-mcp-connector
   
   It's a native Java implementation that integrates directly with Apache Hop's 
APIs and exposes an MCP server through "hop mcp" over STDIO.
   
   Over the past few months, I've been focusing not only on functionality but 
also on making it more reliable and suitable for real ETL environments. The 
current implementation includes:
   
   - Pipeline, workflow, and metadata discovery, inspection, validation, and 
dependency analysis.
   - Native Hop metadata introspection and component schema discovery.
   - Execution, diagnostics, and controlled semantic modifications.
   - Project indexing, security controls, resource limits, and automated 
regression testing.
   
   The latest release is v2.2.6, built against Apache Hop 2.19, with additional 
compatibility testing against the 2.20 snapshot line.
   
   Regarding your proposal, I think extending the existing XML schema 
generation work to "@HopMetadata" classes makes sense. Some of the metadata 
discovery and introspection already implemented in the connector might also be 
useful as a reference for that work.
   
   My original intention was to develop and stabilize the connector 
independently before proposing its integration into Apache Hop, rather than 
introducing something that wasn't sufficiently mature.
   
   Since you've brought up MCP support, I thought this would be a good 
opportunity to share what I've been working on.
   
   I'd appreciate it if you could take a look at the repository when you have a 
chance. I'm open to feedback and would be happy to discuss how the relevant 
parts could eventually be contributed upstream through smaller, manageable PRs.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to