It was originally a mixture of Blaze / Bootstrap / Ionic⦠Used meteor user-status to determine who was actively actually looking at the dashboard (bottom left avatars show activity in real-time like the rest of the dashboard)
Well, after 2 days of Opus 4.6 going at it and a little hand holding⦠itās completely re-written in Typescript, React 19, TailwindCSS 4 and running smoothly! 1.11.1 to 3.5-beta.4 in 2 days!
This tool was my take on real time infrastructure monitoring before grafana was cool
. Plus⦠itās still more real-time than the āauto refreshā you get with tools like grafana/zabbix.
Another successful migration with this workflow:
- Open VS code and start a new workspace
- Go to File - > Add Folder to Workspace
- Add your existing / old meteor project
git clone https://github.com/wreiske/meteor-react-tailwind-prettier-starter project-reactsomewhere on your machine- Go to File - > Add Folder to Workspace ā select project-react
- Open Copilot, make sure itās in āPlanā mode and tell it that there are two projects in the workspace.
Hereās what I used:
There are two projects in the workspace.
statengineandstatengine-react
statengineis an old Meteor 1 application that we need to rewrite into the new modern Meteor 3.4 stack inside ofstatengine-react, which is currently a āTODO example appā that is a āstarterā template for React19, TailwindCSS 4, Typescript, etc.Please come up with an extensive plan for how we can handle a complete migration and cleanup of the original statengine code base into the new format.
After a little while it will have a plan. review it and make any edits you think are required⦠This was my response:
OK. Letās do it. Start the plan. Please make sure you put the entire plan into a markdown file and as you go through your plan, check off each item, so we know where we left off.
Also, iād like to use luxon instead of date-fns.
And then hand hold it with a few ācontinue with the planā¦ā until itās complete! ![]()
This workflow has allowed me to migrate several old projects with very little hand holding and my typical āformat, lint, commit, pushā prompt and copilot instructions to keep code DRY, KISS, etcā¦
YMMV!
Going to do some serious code review before pushing out to a test cluster node and then Iāll check our CPU / Memory / etc old vs new for a few days and see what improvements we get!

