iLearned Self Improving Agent
Use when an OpenClaw, LobeHub, or Hermes assistant encounters user corrections, failed commands or tools, outdated knowledge, missing capabilities, recurring mistakes, or a verified better approach, and before tasks where past lessons may matter. Prefer this skill whenever durable, reusable rules ca
as observed 2026-11-11T20:44:06.643Z- Identifier
ilearned-self-improving-agent- Source
- ClawHub
- Version observed
- 0.1.1
- Source repository
- not published
- Repository observation
- No source repository listed
- First observed here
- 2026-11-06T15:44:01.797Z
- Observations recorded
- 2
- Installs (reported upstream)
- 1
- Weekly downloads (upstream)
- 159
- Declared license
- MIT-0
Observation history
2026-11-11T20:44:06.643Z
Fields that differed: changelog latestVersion summary
| Field | Before | After |
|---|---|---|
changelog |
"Version 0.1.0 introduces a structured workflow for learning from experience and applying verified lessons.\n\n- Adds guidance to search and validate relevant lessons before acting | "- Reworked the guidance into a clearer search–capture–apply–verify–refine workflow, with more detail on rule quality, conflicts, and memory hygiene.\n- Strengthened safety and aut |
latestVersion |
"0.1.0" | "0.1.1" |
summary |
"Use when an OpenClaw, LobeHub, or Hermes assistant encounters user corrections, failed commands or tools, outdated knowledge, missing capabilities, recurring mistakes, or a better | "Use when an OpenClaw, LobeHub, or Hermes assistant encounters user corrections, failed commands or tools, outdated knowledge, missing capabilities, recurring mistakes, or a verifi |
Correction
If you maintain this extension and believe anything above is inaccurate, request a correction. Corrections are published, and disputed entries are marked as disputed while under review.