phase-learning-extraction

Extract decisions, lessons, patterns, and surprises from phase artifacts into LEARNINGS.md.

9|1|Updated Feb 1, 2026
One-click install
npx skills add https://github.com/calcosmic/Aether --skill phase-learning-extraction
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: phase-learning-extraction
Source: https://github.com/calcosmic/Aether/tree/main/.aether/skills-codex/colony/phase-learning-extraction
Command: npx skills add https://github.com/calcosmic/Aether --skill phase-learning-extraction

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Phase learning extraction captures decisions, lessons, patterns, and surprises from completed phase artifacts and produces LEARNINGS.md to feed the colony knowledge base for future phases and milestones.

Core Features & Use Cases

  • Extract Decisions, Lessons, Patterns, Surprises: Collect explicit choices, context, and outcomes from phase work and document them in a structured LEARNINGS file.
  • Generate LEARNINGS.md and update knowledge base: Create a reusable artifact and link it to the colony knowledge repository for cross-phase reference.
  • Milestone and Phase Aggregation: Support milestone-wide themes by aggregating learnings across phases and summarizing key patterns.

Quick Start

Run the phase learning extraction workflow after a phase completes to generate LEARNINGS.md and update the colony knowledge base.

Frequently Asked Questions about phase-learning-extraction

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

FAQPage Schema
How do I extract learnings from completed project phase artifacts?

Phase learning extraction analyzes completed phase artifacts like PLAN.md, IMPLEMENTATION.md, and TEST_RESULTS.md to identify decisions, lessons, and patterns, then compiles them into a structured LEARNINGS.md file for future reference.

What is the best way to document project decisions and surprises for a retrospective?

Documenting project decisions and surprises for a retrospective involves extracting structured choices, context, and outcomes from phase artifacts to generate a LEARNINGS.md file that captures key insights for future milestones.

How do I update a knowledge base with lessons learned from a milestone review?

Updating a knowledge base with lessons learned requires extracting cross-phase patterns from completed milestone artifacts and appending the structured learnings to a colony-knowledge.jsonl data store for future reference.

Do I need specific phase artifacts to generate a LEARNINGS.md file?

Generating a LEARNINGS.md file requires access to completed phase artifacts including PLAN.md, IMPLEMENTATION.md, TEST_RESULTS.md, and REVIEW.md located within the .aether/phases directory structure.

Can I aggregate learnings across multiple phases for milestone-wide themes?

Aggregating learnings across multiple phases supports milestone-wide themes by summarizing key patterns and extracting cross-phase decisions to populate a unified colony knowledge base.

Why does my phase learning extraction need access to a knowledge store?

Phase learning extraction needs access to a colony-knowledge.jsonl knowledge store to enrich the existing repository by appending newly extracted decisions and patterns for cross-phase reference and future success.