# Failure Learning

> Use after a non-trivial, recurring, or expensive failure when a concise evidenced lesson could prevent future repeated investigation.

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

---


# Failure Learning

## Core principle

Save reusable conclusions and recovery, never full debugging transcripts or hidden reasoning.

## Procedure

1. Name the observable symptom.
2. State the evidenced root cause.
3. Record the evidence source and successful recovery.
4. Describe how to detect the pattern earlier.
5. Check stability, reuse value, and future token savings.
6. Store it in failure-patterns or lessons, or reject promotion.

## Required output

Symptom, root cause, evidence, recovery, early signal, and memory destination.

## Limits

One concise entry per reusable pattern.

## Common mistakes

Do not save temporary errors, speculative diagnoses, giant output, or raw chain-of-thought.

## Stop condition

Stop when the reusable lesson is stored concisely or judged too temporary.