learning-patterns

Commit staged behavioral learnings to MCP Memory and update CLAUDE.md.

Updated Mar 9, 2026
One-click install
npx skills add https://github.com/shakedarazi/session-intelligence --skill learning-patterns
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: learning-patterns
Source: https://github.com/shakedarazi/session-intelligence/tree/main/skills/learning-patterns
Command: npx skills add https://github.com/shakedarazi/session-intelligence --skill learning-patterns

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps you manage and integrate the insights your AI assistant gains about your working habits, turning passive observations into actionable improvements.

Core Features & Use Cases

  • Commit Learnings: Push observed behavioral patterns to a central memory for long-term storage and analysis.
  • Review Patterns: Get a clear overview of learned preferences, workflows, and gaps, sorted by confidence.
  • Graduate Patterns: Promote well-established patterns into your primary configuration (CLAUDE.md) for consistent application.
  • Prune Patterns: Clean up outdated or incorrect observations to maintain an accurate understanding of your workflow.
  • Use Case: After the AI has been observing your work for a while, you can use this skill to review what it has learned about your priorities and then graduate the most consistent patterns into your CLAUDE.md so the AI can proactively adapt to them.

Quick Start

Ask the learning patterns skill to push your staged learnings to MCP Memory.

Frequently Asked Questions about learning-patterns

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

FAQPage Schema
How do I commit staged learnings and behavioral patterns to MCP Memory?

Behavioral pattern management involves capturing session context observations, committing them to a graph database, and promoting proven workflows into CLAUDE.md to proactively adapt AI assistant behavior to your habits.

How do I promote proven workflow patterns into CLAUDE.md?

You promote proven patterns by reviewing learned preferences sorted by confidence, then graduating well-established behavioral observations from MCP Memory into CLAUDE.md for consistent AI application across sessions.

Can I prune stale observations and outdated patterns from memory?

Yes, you can prune stale observations and outdated patterns to clean up incorrect behavioral data, ensuring the graph database and CLAUDE.md maintain an accurate understanding of your current workflow.

Does this pattern learning workflow require a graph database dependency?

Yes, the workflow requires a graph database dependency, specifically interacting with MCP Memory via functions like search_nodes and create_entities to store and retrieve behavioral observations.