aget-record-lesson

Capture session lessons as templated markdown documents with classification and indexing.

1|Updated Oct 4, 2025

Search Tags:#git #templates #markdown #indexing #classification #knowledge-capture #lessons-learned

Ready to run

npx skills add https://github.com/gmelli/public-OpenAI-DeepResearch-aget --skill aget-record-lesson-gmelli

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: aget-record-lesson
Source: https://github.com/gmelli/public-OpenAI-DeepResearch-aget/tree/main/.claude/skills/aget-record-lesson
Command: npx skills add https://github.com/gmelli/public-OpenAI-DeepResearch-aget --skill aget-record-lesson-gmelli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Capture transient session observations, errors, and improvements as persistent, searchable, committable lesson documents so important learnings survive context windows and session boundaries and can be reused across projects and agents.

Core Features & Use Cases

Classification: Guides users to classify each lesson as Framework (reusable across AGETs) or Domain (specific to the principal) before writing.
Template-driven creation: Produces standardized markdown L-docs in .aget/evolution or knowledge/patterns with metadata and traceability fields.
Indexing and traceability: Reads and updates .aget/evolution/index.json for Framework lessons to maintain sequential IDs and discoverability.
Use Case: After encountering a reproducible context loss pattern, record the insight as a Framework lesson with a numbered L-document so other agents can adopt the mitigation.

Quick Start

Record the lesson observed in the last session about context anchoring as a Framework titled "Context-Anchoring Blindness" with a one-paragraph summary and save it to the evolution index.

Frequently Asked Questions about aget-record-lesson

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

FAQPage Schema

How do I capture session lessons as searchable markdown documents for reuse?▼

To capture session lessons as searchable markdown documents, you classify the insight as Framework or Domain and generate a templated markdown file with metadata in .aget/evolution or knowledge/patterns for traceability and reuse.

How does indexing work for lessons learned in agent development workflows?▼

Indexing for lessons learned involves reading and updating .aget/evolution/index.json to assign sequential IDs to Framework lessons, maintaining discoverability and traceability across agent sessions.

What's the best way to classify knowledge capture insights as Framework or Domain?▼

Classifying knowledge capture insights as Framework or Domain depends on reusability: Framework lessons apply across AGETs and require index updates, while Domain lessons are specific to the principal and stored without indexing.

Can I use markdown templates for knowledge capture without git dependencies?▼

Yes, you can use markdown templates for knowledge capture without git dependencies because the Skill only requires explicit user classification and file creation in local directories, though committing to git improves traceability.

When do I need to update index.json during the lesson recording process?▼

You need to update index.json during the lesson recording process only when adding Framework lessons, ensuring sequential IDs are maintained for cross-project agent discovery and reuse.

Why does classifying lessons as Domain instead of Framework affect storage location?▼

Classifying lessons as Domain instead of Framework affects storage because Domain lessons are stored in knowledge/patterns without index updates, whereas Framework lessons require .aget/evolution and index.json modifications for traceability.