Learning Capture
A framework for recognizing and formalizing valuable patterns from work sessions into reusable skills with conservative, ROI-focused pattern recognition.
This skill is designed for skill developers and power users who want to systematically identify and capture reusable patterns from their work sessions.
Skill Structure
The repository provides a comprehensive framework for pattern recognition and skill creation:
Main Files:
- SKILL.md - Core learning capture framework and decision criteria
Resource Directories:
- references/ - Supporting materials including:
decision-examples.md- Real-world examples of capture decisionsskill-templates.md- Templates for creating new skills
Core Purpose
The Learning Capture framework aims to recognize valuable patterns that will save significant context window tokens through frequent reuse, specifically targeting patterns that will be reused 10+ times and save 500+ tokens per reuse.
This systematic approach helps build a knowledge base of truly valuable patterns while avoiding noise from one-off or overly specific solutions.
Five Recognition Triggers
The framework monitors for five types of learning moments:
1. Novel Problem-Solving Approaches
Creative, generalizable solutions to complex problems that extend beyond the specific instance and could apply to similar challenges.
2. Repeated Patterns
Tasks requested 2-3 times with consistent structure emerging, indicating a pattern worth formalizing for future reuse.
3. Domain-Specific Knowledge
Company processes and terminology accumulated across multiple conversations, representing institutional knowledge worth capturing.
4. Effective Reasoning Patterns
Thinking structures that consistently produce superior results and could benefit other problem-solving scenarios.
5. Workflow Optimizations
Efficient tool chains combining 3+ distinct steps that create a comprehensive, repeatable outcome.
Key Decision Criteria
Before offering pattern capture, all three conditions must be met:
Required Criteria:
- High Confidence (>95%): Pattern will be reused 10+ times across future conversations
- Strong Reusability Signals: Clear evidence of pattern's ongoing value already present
- No Redundancy: Pattern doesn't overlap with existing captured capabilities
The framework explicitly warns against offering capture on:
- First instances of a pattern
- Highly context-specific solutions
- Ambiguous scenarios where reusability is unclear
Conservative Approach
The guidance emphasizes being "conservative by default":
"Better to capture 80% of truly valuable patterns than create noise in the knowledge base."
This philosophy prioritizes quality over quantity, ensuring that captured patterns genuinely provide ROI through frequent reuse and significant token savings.
Repository Resources
The repository includes detailed references and templates for:
- Decision-making examples showing when to capture vs. skip
- Skill templates for structuring captured patterns
- Best practices for pattern recognition and formalization
Visit the Learning Capture repository for complete documentation and reference materials.
About This Skill
This skill was created by Nate Jones as part of his comprehensive Nate's Substack Skills collection. Learn more about Nate's work at Nate's Newsletter.
Explore the full collection to discover all 10+ skills designed to enhance your Claude workflows!
Framework for recognizing and formalizing valuable patterns from work sessions into reusable skills with conservative, ROI-focused pattern recognition.