continuous-learning-v2

Observe Claude Code sessions and extract atomic patterns with confidence scoring.

1|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/Michae2xl/claude-skills-michael --skill continuous-learning-v2-michae2xl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: continuous-learning-v2
Source: https://github.com/Michae2xl/claude-skills-michael/tree/main/skills/continuous-learning-v2
Command: npx skills add https://github.com/Michae2xl/claude-skills-michael --skill continuous-learning-v2-michae2xl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python, yaml, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill turns your Claude Code sessions into reusable knowledge by observing sessions, creating atomic instincts with confidence scoring, and evolving them into skills/commands/agents.

Core Features & Use Cases

  • Automated Learning: Observes Claude Code sessions and creates instincts from the patterns.
  • Confidence Scoring: Provides confidence scoring for each instinct based on observation frequency.
  • Instinct Evolution: Evolves instincts into full skills, commands, or agents.
  • Project-Scope Instincts: Prevents cross-project contamination with project-scoped instincts.
  • Use Case: Use this Skill to automatically extract and evolve patterns from your codebase into reusable code snippets or tools.

Quick Start

To enable observation hooks for the 'continuous-learning-v2' skill, add the following to your '.claude/settings.json':

{
  "hooks": {
    "PreToolUse": [{
      "matcher": "*",
      "hooks": [{
        "type": "command",
        "command": "${CLAUDE_PLUGIN_ROOT}/skills/continuous-learning-v2/hooks/observe.sh"
      }]
    }],
    "PostToolUse": [{
      "matcher": "*",
      "hooks": [{
        "type": "command",
        "command": "${CLAUDE_PLUGIN_ROOT}/skills/continuous-learning-v2/hooks/observe.sh"
      }]
    }]
  }
}

Frequently Asked Questions about continuous-learning-v2

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

FAQPage Schema
How do I extract reusable code patterns from Claude Code sessions?

To extract reusable code patterns from Claude Code sessions, you configure observation hooks in your settings to monitor tool usage. The Skill analyzes these interactions to create atomic instincts with confidence scoring based on observation frequency.

What is automated learning from codebase interactions?

Automated learning from codebase interactions is the process of observing development sessions to identify and extract recurring patterns. This Skill captures these patterns as atomic instincts and evolves them into reusable skills, commands, or agents.

How do I prevent cross-project contamination when extracting code patterns?

You prevent cross-project contamination when extracting code patterns by using project-scoped instincts. This Skill isolates learned patterns within their specific project boundaries to ensure extracted knowledge remains relevant to its original context.

Do I need Python to analyze codebase patterns and evolve them into tools?

Yes, you need Python to analyze codebase patterns and evolve them into tools. The Skill requires Python to perform instinct analysis, confidence scoring, and the subsequent evolution of those instincts into functional skills or commands.

How do I set up observation hooks for continuous learning in Claude Code?

To set up observation hooks for continuous learning in Claude Code, you add PreToolUse and PostToolUse hook configurations to your '.claude/settings.json' file. These hooks execute the observe.sh script to monitor session activity.

What's the best way to evolve raw development patterns into reusable agents?

The best way to evolve raw development patterns into reusable agents is through confidence scoring and pattern recognition. This Skill observes sessions, scores the frequency of atomic instincts, and evolves high-confidence patterns into autonomous agents.