Add shadow-tutor skill#466
Open
lzfxxx wants to merge 1 commit into
Open
Conversation
A post-session programming tutor: after an AI-assisted coding session it teaches why each decision was made (including rejected options), bound to the session's real diffs, then drills the user with test-graded forced-recall exercises. Self-contained (SKILL.md + METHODOLOGY.md + scripts/), MIT licensed. Source and eval harness: https://github.com/lzfxxx/shadow-tutor
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Adds shadow-tutor under
skills/.curated/shadow-tutor/— a post-session programming tutor.After an AI-assisted coding session, it produces a short, evidence-backed review of why the agent made each decision (including the options it rejected), every point bound to the session's real diffs, then walks the user through forced-recall exercises graded against the project's own tests. It runs in-session on the user's existing model (no API key), and is format-identical for Codex and Claude Code.
Why
AI coding agents make it easy to ship code you don't understand — learning lives in the struggle the agent removes. This skill re-inserts that cognitive friction after the work, so developers actually learn instead of just shipping. It's built with Codex's own skills paradigm.
Contents
Self-contained skill directory:
SKILL.md— routing metadata + entry instructionsMETHODOLOGY.md— the teaching methodology (evidence-first selection, why-over-what, graded exercises, structured closeout)scripts/knowledge.mjs— per-user knowledge profile (schema-validated, atomic writes)LICENSE.txt— MITNotes
.curated/; happy to move it if there's a preferred location/process for community submissions.