I admittedly haven't tried this approach at work yet but at home while working on a side project, I'll make a new feature branch and give CLAUDE a prompt about what the feature is with as much detail as possible. i then have it generate a CLAUDE-feature.md and place an implementation plan along with any supporting information (things we have access to in the codebase, etc.).
i'll then prompt it for more based on if my interpretation of the file is missing anything or has confusing instructions or details.
usually in-between larger prompts I'll do a full /reset rather than /compact, have it reference the doc, and then iterate some more.
once it's time to try implementing I do one more /reset, then go phase by phase of the plan in increments /reset-ing between each and having it update the doc with its progress.
generally works well enough but not sure i'd trust it at work.
i'll then prompt it for more based on if my interpretation of the file is missing anything or has confusing instructions or details.
usually in-between larger prompts I'll do a full /reset rather than /compact, have it reference the doc, and then iterate some more.
once it's time to try implementing I do one more /reset, then go phase by phase of the plan in increments /reset-ing between each and having it update the doc with its progress.
generally works well enough but not sure i'd trust it at work.