The way I do it is with an agents.md prompt (though the extension provides a skill). Basically, if a Kanban item is created and referenced, ai uses that to log changes, decisions, and proposed commit messages etc. If there isn't a Kanban item, it creates one and does the same.
I am using another VSCode Kanban extension. Very similar workflow to this one. I am very happy with it, it solved many issues I am having with context.
This was me until I discovered Logseq. I still use Workflowy to collaborate with others, but combining notes and todos with a powerful query system is what makes Logseq my goto note and todo management tool.
I am in love with wails. having python and JS background with no go experience. I pm'ed Ai agents to create a fairly complex desktop app for my own use. it is day and night in terms of performance compared to lightest electron app.
This is a great read. I’ve also read a similar paper discussing bad luck with rare events, like lightning strikes and plane crashes, happening to the same person more than once, and the statistical side of it. I think about that paper from time to time and try to find it on Google, but haven’t had any luck.