# Feedback Personalizer

> Generates per-student actionable feedback mapped to rubric criteria and evidence, with concrete next-step recommendations.

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

---


# Feedback Personalizer

## Goal
Turn scores and evidence into concise, actionable learning feedback.

## Inputs
- `rubric.md`
- Per-student evidence files
- Scores and criterion-level outcomes

## Produce
For each learner, write `feedback/<student-id>.md` with:
- strengths tied to rubric criteria
- weaknesses tied to explicit evidence
- priority fixes (short term)
- practice plan (medium term)

Also write `feedback-summary.json` with cohort-level patterns.

## Rules
- No generic praise/criticism without criterion reference.
- Separate factual evidence from interpretation.
- Keep tone professional and improvement-oriented.
- Avoid disclosing solution keys beyond allowed policy.