Highlights of the Mg 0.6.2 Release
We’re excited to announce the release of Mg 0.6.2, the first version to introduce integration with AI agents through Jupyter AI and Goose.
As previously announced, we have been actively exploring the use of LLMs for Model-based Human-Machine Collaboration (MHMC). Integrating AI agents is an important step toward streamlining engineering tasks, including those involving Mg processes and operations.
This is just the beginning. We plan to further incorporate our expertise into these AI-powered solutions, including AOP (Algebraic Ontology Projection), to enhance engineering productivity and collaboration.
This release also upgrades the SysML v2 Pilot Implementation to version 0.59.0 (2026-04), aligning Mg with the latest standard updates.
Known Issues
Due to the limitation of the current Jupyter AI, MCP servers of multiple Mg instances do not work well. AI agents can use the MCP servers of the first instance of Mg. Note that Mg also hosts its own MCP server and its port is currently fixed to 13297.
Looking Ahead
We continue to actively support MgDNG (DOORS Next integration) and are focusing on AI agent optimization and expect to report the results in a timely manner. Stay tuned!