agentskills.codes

Use continuously during work to capture mistakes, discoveries, corrections, and non-obvious repo learnings.

Install

mkdir -p .claude/skills/napkin-engraphcode && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/13864" && unzip -o skill.zip -d .claude/skills/napkin-engraphcode && rm skill.zip

Installs to .claude/skills/napkin-engraphcode

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Use continuously during work to capture mistakes, discoveries, corrections, and non-obvious repo learnings.
107 charsno explicit “when” trigger

About this skill

Napkin

Goal

Keep the capture stage of the knowledge flow active during every session.

Workflow

  1. At session start, read .agent/memory/distilled.md if it exists.
  2. Then read .agent/memory/napkin.md if it exists.
  3. Write to the napkin whenever you learn something worth preserving:
    • tool surprises
    • user corrections
    • mistakes and recoveries
    • architectural discoveries
  4. Be specific and actionable.
  5. When the napkin grows beyond roughly 500 lines, use the distillation skill.

Search skills

Search the agent skills registry