continuous-learning

Capture recurring patterns and architectural decisions into explicit standards.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/RogerioSobrinho/codeme-copilot --skill continuous-learning-rogeriosobrinho
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: continuous-learning
Source: https://github.com/RogerioSobrinho/codeme-copilot/tree/main/skills/continuous-learning
Command: npx skills add https://github.com/RogerioSobrinho/codeme-copilot --skill continuous-learning-rogeriosobrinho

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Load when capturing a new architectural decision (ADR), documenting a discovered pattern or anti-pattern after solving a complex problem, applying the three-time-rule (promote to standard after solving the same problem 3 times), tracking an instinct with a confidence score, promoting a recurring instinct to an explicit standard, or at the end of a session to preserve reusable solutions. Triggers on: "document this", "remember this pattern", "write an ADR", "add to our standards", "this keeps coming up", "track this instinct", "I have a feeling about this approach".

Core Features & Use Cases

  • Instinct lifecycle with confidence scoring and the ability to promote instincts to explicit patterns or standards.
  • Three-time rule for pattern promotion, ensuring repeated success before generalizing.
  • ADR templates and end-of-session review protocol to preserve reusable solutions and drive governance.

Quick Start

Document a new ADR or pattern by creating or updating SKILL.md with the instinct, evidence log, and promotion status.

Frequently Asked Questions about continuous-learning

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

FAQPage Schema
How do I document an architectural decision record to capture engineering patterns?

To document an architectural decision record (ADR), use provided ADR templates to formalize insights and recurring patterns into explicit standards. This captures architectural decisions and builds a reusable knowledge base during post-solution reflections or cross-project reviews.

What is the three-time rule for promoting engineering instincts to standards?

The three-time rule for pattern promotion requires solving the same problem successfully three times before generalizing a recurring instinct into an explicit standard. This ensures repeated success validates the pattern before formalizing it into your engineering best practices.

How does tracking an instinct with a confidence score work for knowledge management?

Tracking an instinct with a confidence score implements an instinct lifecycle where informal engineering feelings are logged and scored. As evidence accumulates across sessions, high-confidence instincts become eligible for promotion into formal, documented patterns or standards.

When should I capture recurring patterns for continuous learning during a project?

Capture recurring patterns for continuous learning when writing an ADR, documenting a discovered anti-pattern after solving complex problems, or during end-of-session archival. Triggers include phrases like 'remember this pattern' or 'this keeps coming up' to preserve reusable solutions.

What is the best way to preserve reusable solutions at the end of a development session?

The best way to preserve reusable solutions at the end of a session is using an end-of-session review protocol to archive discovered patterns and instincts. This updates your knowledge base by formalizing evidence logs and promoting validated instincts into standards.

Can I use continuous learning for cross-project reviews and knowledge management without external dependencies?

Yes, continuous learning operates without external dependencies, using internal ADR templates and evidence logs for cross-project reviews. It archives reusable solutions directly within your project structure to build a standalone, reusable knowledge base.