skill-debrief

Convert AI agent session feedback into approved SKILL.md edits.

450|36|Updated Jan 3, 2026

Search Tags:#documentation #retrospective #ai agent #skill improvement #prompt updates #workflow fixes

Ready to run

npx skills add https://github.com/computerlovetech/agr --skill skill-debrief

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: skill-debrief
Source: https://github.com/computerlovetech/agr/tree/main/skills/skill-debrief
Command: npx skills add https://github.com/computerlovetech/agr --skill skill-debrief

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you capture what went wrong (or what went right) during an AI agent session and translate that experience into concrete improvements to the skill that powered it.

Core Features & Use Cases

Debrief-by-conversation: Lets you share session feedback in any form and distills it into actionable changes, including trigger updates and workflow fixes.
Edit-and-reinstall for in-repo skills: Applies approved edits directly to skills/<name>/SKILL.md (and related files) and guides you through committing and upgrading via agr upgrade.
Upstream feedback for remote skills: For skills installed from upstream repos, helps you prepare an issue or fork workflow so improvements can be adopted over time.

Quick Start

Ask your AI tool to debrief the skill you just used by saying: "debrief the skill that handled my task and improve it based on what happened."

Frequently Asked Questions about skill-debrief

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

FAQPage Schema

How do I improve an AI agent skill after a session went wrong?▼

To improve an AI agent skill after a failed session, you can debrief the experience to capture workflow errors and translate them into concrete edits for your SKILL.md file and related documentation.

How do I fix prompt triggers that over-fired or under-fired during a session?▼

Fixing prompt triggers that over-fired or under-fired involves debriefing the session specifics to propose scoped changes, updating the skill boundaries, and reinstalling the modified SKILL.md.

What is the best way to document retrospective feedback for AI workflows?▼

Documenting retrospective feedback for AI workflows requires capturing what went wrong or right during the session and converting that feedback into actionable prompt updates and workflow fixes.

Can I apply skill improvements to remote upstream repos?▼

Yes, for remote upstream repos, you can apply skill improvements by preparing an issue or fork workflow so your feedback can be adopted upstream over time.

Does updating an in-repo skill require manual reinstallation?▼

Updating an in-repo skill applies approved edits directly to skills/<name>/SKILL.md and guides you through committing and upgrading the installation automatically via agr upgrade.

When should I not use a session debrief for skill improvement?▼

You should avoid a session debrief when you lack specific session feedback or cannot explicitly approve scoped changes, as the process requires listening for session specifics to propose targeted updates.