# Self Improving Agent

> Closed-loop learning system — captures errors, corrections, and discoveries into structured logs, promotes recurring patterns to project memory, and extracts reusable skills

- Skill: `memi-design/self-improving-agent` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add memi-design/self-improving-agent`
- Raw SKILL.md: https://api.skillmd.com/api/skills/memi-design/self-improving-agent/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: memi-design (https://skillmd.com/u/memi-design)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/memi-design/self-improving-agent

---


# Improve an agent from evidence

Read [the closed-loop learning guide](references/guide.md). Capture concrete failures, corrections, and successful recovery steps without storing secrets or unnecessary personal data. Promote only recurring, validated patterns; keep temporary session context out of durable team guidance. Prefer updates to an existing project document or skill over creating overlapping memory. Verify that promoted guidance changes a realistic future task.