About continuous-learning
A Claude Code skill that reviews each completed session for reusable patterns and saves selected findings as learned skills. Its settings control session length, extraction threshold, approval, storage location, and pattern types.
Key features
- 01End-of-session evaluation hook
- 02Reusable pattern detection
- 03Learned skill extraction
- 04Configurable approval and thresholds
Use cases
- 01Saving recurring solutions to errors
- 02Capturing debugging methods and workarounds
- 03Recording project-specific conventions
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ worldflowai/ everything-claude-code --skill continuous-learningAfter installing, type/continuous-learningin Claude Code to confirm it loaded.
Bundled filesconfig.jsonevaluate-session.sh
Details
From the project READMEContinuous Learning Skill
Automatically evaluates Claude Code sessions on end to extract reusable patterns that can be saved as learned skills.
How It Works
This skill runs as a Stop hook at the end of each session:
- Session Evaluation: Checks if session has enough messages (default: 10+)
- Pattern Detection: Identifies extractable patterns from the session
- Skill Extraction: Saves useful patterns to
~/.claude/skills/learned/
Configuration
Edit config.json to customize:
{
"min_session_length": 10,
"extraction_threshold": "medium",
"auto_approve": false,
"learned_skills_path": "~/.claude/skills/learned/",
"patterns_to_detect": [
"error_resolution",
"user_corrections",
"workarounds",
"debugging_techniques",
"project_specific"
],
"ignore_patterns": [
"simple_typos",
"one_time_fixes",
"external_api_issues"
]
}
Pattern Types
| Pattern | Description |
|---|---|
error_resolution | How specific errors were resolved |
user_corrections | Patterns from user corrections |
workarounds | Solutions to framework/library quirks |
debugging_techniques | Effective debugging approaches |
project_specific | Project-specific conventions |
Hook Setup
Add to your ~/.claude/settings.json:
{
"hooks": {
"Stop": [{
"matcher": "*",
"hooks": [{
"type": "command",
"command": "~/.claude/skills/continuous-learning/evaluate-session.sh"
}]
}]
}
}
Why Stop Hook?
- Lightweight: Runs once at session end
- Non-blocking: Doesn't add latency to every message
- Complete context: Has access to full session transcript
Related
- The Longform Guide - Section on continuous learning
/learncommand - Manual pattern extraction mid-session
Comments
Markdown supported