continuous-learning

Analyze Claude Code sessions and store reusable patterns as learned skills.

2|1|Updated Oct 4, 2025
One-click install
npx skills add https://github.com/andrew-starosciak/deep-algo --skill continuous-learning-andrew-starosciak
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: continuous-learning
Source: https://github.com/andrew-starosciak/deep-algo/tree/main/.claude/docs/zh-TW/skills/continuous-learning
Command: npx skills add https://github.com/andrew-starosciak/deep-algo --skill continuous-learning-andrew-starosciak

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automatically analyzes Claude Code work sessions to extract reusable patterns and save them as learned skills for future use.

Core Features & Use Cases

  • Stop-hook based learning: Runs at the end of each Claude Code session to evaluate activity.
  • Pattern detection: Identifies reusable patterns such as error_resolution, user_corrections, workarounds, debugging_techniques, and project_specific.
  • Skill persistence: Stores extracted patterns to ~/.claude/skills/learned/.

Quick Start

Ensure the continuous-learning skill is available at ~/.claude/skills/continuous-learning, then add a Stop hook in your Claude Settings to call evaluate-session.sh after each session. Upon the next session completion, the script collects data, extracts patterns, and writes learned skills to ~/.claude/skills/learned/.

Frequently Asked Questions about continuous-learning

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

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

You can automatically extract reusable patterns from Claude Code sessions by configuring a Stop hook to run an evaluation script after each session, detecting predefined patterns and saving them as learned skills.

What types of debugging patterns can be automatically saved as learned skills?

The system detects and saves debugging patterns such as error resolution, user corrections, workarounds, debugging techniques, and project-specific patterns as learned skills for future use.

How do I configure continuous learning hooks to analyze session activity?

Configure continuous learning by setting parameters like minimum session length, extraction threshold, auto approve, learned skills path, and patterns to detect in a config.json file, then link the hook script.

When do I need a Stop hook for session analysis in Claude Code?

You need a Stop hook for session analysis when you want to evaluate activity and capture workarounds automatically at the end of each Claude Code session, rather than manually reviewing logs.

Does Claude Code support saving extracted patterns directly to a local directory?

Yes, Claude Code supports saving extracted patterns directly to a local directory, specifically storing the generated learned skills in the ~/.claude/skills/learned/ path.

What are the limitations of using Stop hooks to capture user corrections?

A limitation of using Stop hooks is that pattern extraction relies on configured thresholds and minimum session lengths, meaning very short sessions or patterns below the extraction threshold will not be saved.