session-retrospective

Capture session corrections and lessons into .ai/memory/lessons.md.

1.5k|257|Updated Nov 19, 2018

Search Tags:#retrospective #memory #session-management #lessons #AI-agents

Ready to run

npx skills add https://github.com/adobe/spectrum-web-components --skill session-retrospective-adobe

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: session-retrospective
Source: https://github.com/adobe/spectrum-web-components/tree/main/.ai/skills/session-retrospective
Command: npx skills add https://github.com/adobe/spectrum-web-components --skill session-retrospective-adobe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Document lessons learned after completing work, especially when the user corrected planning documents or implementation. Creates and maintains a persistent lessons file in .ai/memory/ that future agents read at session start.

Core Features & Use Cases

Systematically capture corrections and surprises from a session.
Organize lessons by category and provide quick reference for future tasks.
Use this to accelerate learning and reduce repeating the same mistakes.

Quick Start

Create or update the memory file with the latest corrections from this session so future agents can read it at startup.

Frequently Asked Questions about session-retrospective

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

FAQPage Schema

How do I capture lessons learned from an AI agent session?▼

Capture lessons learned from an AI agent session by automatically extracting corrections and writing them to a persistent memory file. This ensures that future agents can read the lessons at session start to avoid repeating past mistakes.

How does persistent memory work for AI agents across multiple sessions?▼

Persistent memory works by storing session corrections and plan changes in a structured markdown file under the .ai/memory/ directory. Future agents access this file at startup to retrieve categorized lessons from previous interactions.

What is the best way to stop AI agents from repeating the same mistakes?▼

The best way to stop AI agents from repeating mistakes is to systematically document corrections and surprises in a persistent lessons file. This creates a structured knowledge base that future agents read before beginning new tasks.

How do I structure session retrospective files for future AI agents?▼

Structure session retrospective files by writing to .ai/memory/lessons.md and categorizing entries according to defined topics. This workflow updates existing lessons and provides a quick reference for future tasks.

When are session corrections applied to the persistent memory file?▼

Session corrections are applied to the persistent memory file during ongoing sessions and immediately after plan changes. This ensures that lessons are captured in real-time and are available to future agents at session start.