Diogenesoftoronto/keating / pi/skills/adaptive-teaching
adaptive-teaching
VerifiedUse when the user wants to learn, be taught, be quizzed, or understand an idea across any domain — science, math, philosophy, code, law, politics, psychology, medicine, arts, or history — with active scaffolding.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.0 KB |
SKILL.md
View on GitHub---
name: adaptive-teaching
description: Use when the user wants to learn, be taught, be quizzed, or understand an idea across any domain — science, math, philosophy, code, law, politics, psychology, medicine, arts, or history — with active scaffolding.
---
# Adaptive Teaching
Use this skill when the user asks to learn a concept rather than merely receive a summary.
## Teaching Contract
1. Diagnose first if the learner state is unknown. Check `.keating/state/learner.json` for prior coverage.
2. Teach intuition before formalism, but do not omit formalism for math, science, or code.
3. Surface one misconception early.
4. Use ephemeral OpenUI `question` interactions for retrieval and reconstruction, not just exposition. Stop and wait after each interaction.
5. End with transfer, reflection, or a next challenge.
6. Verify factual claims before teaching. If `.keating/outputs/verifications/` has no checklist for the topic, run `/verify <topic>` first.
## Keating Artifacts
- Author learner-facing questions, plans, concept maps, quizzes, notes, and other teaching surfaces as OpenUI.
- Use `/plan <topic>` or `/map <topic>` only when a persisted filesystem artifact is explicitly useful; do not use them merely to render a card.
- Run `/animate <topic>` yourself when a persisted animation would help — do not ask the user to do it.
- Run `/verify <topic>` yourself before teaching factual claims — do not ask the user to do it.
- Run `/bench [topic]` and `/evolve [topic]` yourself when improvement is needed — do not ask the user to do it.
- Run `/feedback <up|down|confused> [topic]` yourself after sessions to record outcomes.
## Red Flags
- Do not drown the learner in formal language before a concrete hook exists.
- Do not act as though the synthetic benchmark proves pedagogical truth.
- Do not confuse a correct answer with stable understanding; probe transfer.
- Do not present unverified factual claims as settled truth. If verification artifacts exist, consult them. If claims are flagged as unconfirmed, hedge appropriately.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |