leave-advice

Convert post-task learnings into concise, actionable advice for future agents.

Updated Feb 25, 2026
One-click install
npx skills add https://github.com/mikewolfd/formspec --skill leave-advice-mikewolfd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: leave-advice
Source: https://github.com/mikewolfd/formspec/tree/main/.claude-plugin/skills/leave-advice
Command: npx skills add https://github.com/mikewolfd/formspec --skill leave-advice-mikewolfd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

End-of-task reflection tool to capture lessons learned and advice to guide future agents, reducing repeat mistakes and friction.

Core Features & Use Cases

  • Facilitates a concise post-task guidance note.
  • Encourages documenting actionable advice for future runs and agents.
  • Use Case: after completing a project task, generate a short "leave advice" entry to improve next iteration or agent.

Quick Start

Summarize the key lesson learned from this task into a brief, shareable advice note.

Frequently Asked Questions about leave-advice

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I capture post-task learnings for future AI agents?

To capture post-task learnings, you can generate a concise, actionable advice note that documents key lessons learned, reducing repeat mistakes and friction for future agents. This end-of-task reflection converts insights into shareable guidance.

What is the best way to document end-of-task reflection for agent memory?

The best way to document end-of-task reflection is to summarize the key lesson learned into a brief, shareable advice note. This captures knowledge immediately after a project task, ensuring clear guidance for the next iteration or future agent runs.

Can I use post-task reflection notes across common agent workflows?

Yes, you can use post-task reflection across common workflows. The scope is designed for Claude-like agents, ensuring the captured advice adheres to safety, clarity, and applicability criteria regardless of the specific workflow completed.

How do I write actionable advice for future agents after completing a task?

To write actionable advice for future agents, identify post-task learnings and convert them into concise guidance. This ensures the documented advice reduces friction and prevents repeat mistakes in subsequent runs or agent iterations.

Does generating leave advice require any specific dependencies or components?

Generating leave advice requires no specific dependencies or components. It is a basic implementation tool that facilitates a concise post-task guidance note directly from your summarized key lessons without needing external modules.

Why should I capture lessons learned before finishing an agent task?

You should capture lessons learned before finishing an agent task to reduce repeat mistakes and friction. Documenting this knowledge creates a brief, actionable guidance note that improves the efficiency and safety of future agent runs.