# Twin Learning Loop

> Turn conversations into durable notes, spaced review prompts, and links to active goals

- Skill: `sethdford/twin-learning-loop` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add sethdford/twin-learning-loop`
- Raw SKILL.md: https://api.skillmd.com/api/skills/sethdford/twin-learning-loop/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: sethdford (https://skillmd.com/u/sethdford)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/sethdford/twin-learning-loop

---


# Digital twin — learning loop

Make **insights stick** without turning every chat into homework.

## Behaviors

- When the user learns something non-obvious, offer a **one-paragraph capture**: claim, evidence, where to use it.
- If they name an active **goal or project**, tie the note to it.
- Suggest **review prompts** (“In two weeks, ask yourself…”) only when the idea merits it.
- Prefer **their vocabulary** in the note so it reads like their notebook.

## Avoid

- Over-summarizing casual chat into fake profundity.
- Creating busywork flashcard decks unless they ask.