Offer or run 1-3 short retrieval questions about work just completed (why it works, what-if, where to look if it broke), grade the user's answers kindly, and save the reusable rule as a lesson that comes back when a similar situation appears. Use when sharp's…
Private reliance mirror. Summarizes how the user worked with AI over the last week (task types, coaching skipped, one-way doors, how well lessons transfer, lessons that keep slipping) and suggests one concrete habit for next week.
Monthly AI-off checkup that measures real learning (retention, transfer, explanation) with confidence ratings, so growth is measured rather than felt. About 10 minutes.
Decision journal and retrospectives. Lists one-way door decisions the user made, and replays ones that are due for review (what we expected vs what happened) to build calibrated judgment.
Feynman technique. The user explains a topic, change or decision in their own words; Claude finds the gaps, errors and fuzzy parts without rewriting the explanation for them. With no topic, uses the current PR or branch.
Personal growth plan. Shows the user's growth map (topics they're learning, what's mature, what keeps slipping), sets focus areas for deliberate practice, sets a mentor, and drafts an optional note to share with that mentor.
Learn from a PR, a branch or a recent change (the default when no topic is given), or learn a topic fast. For a change, it first gathers the full context (commits, PR description, diff), then explains it in plain words so the user can handle similar changes…
Practise past lessons by applying them to NEW scenarios (transfer drills), not by reciting answers. One lesson at a time, a few minutes total; spacing grows as the user shows they can apply it.