How is pattern confidence calculated?

Confidence is a weighted score combining occurrence frequency, severity of the findings, the recency of the observations, and the quality of the data source.

What do the /evolve and /patterns commands do?

The /evolve command triggers an analysis of new findings to update runbooks, while /patterns displays a registry of detected patterns and their statistical frequency.

When are patterns automatically promoted to runbooks?

Patterns are promoted when they meet four criteria: at least 3 occurrences, a confidence score of 0.6 or higher, a clear fix template, and no existing duplicates.

Does the Learning Engine work across multiple services?

Yes, it performs cross-service matching by checking for similar code structures, shared dependencies, and historical correlations across different parts of the project.

How does the Learning Engine identify recurring code patterns?

The engine analyzes structured data from review findings, session learnings, and user corrections, grouping similar structures to calculate occurrence frequency and confidence.

Learning Engine for Claude Code

Name: Learning Engine for Claude Code
Author: datum-cloud

bydatum-cloud

•

학습 및 문서화

Automates pattern extraction and knowledge accumulation from code reviews and session findings to continuously improve development workflows.

The Learning Engine transforms raw code review findings, session learnings, and user corrections into actionable knowledge by automatically detecting recurring patterns across services. It calculates confidence scores based on frequency and severity, generates trend analysis to track code health, and automatically promotes high-confidence patterns into documentation and runbooks. This skill ensures that Claude and developers stay aligned with evolving best practices by turning every code interaction into a permanent improvement for the entire project.

주요 기능

01Multi-factor confidence scoring based on frequency, severity, and recency

02Cross-service pattern matching to identify systemic issues across repositories

03Automated runbook generation and promotion for high-confidence insights

04Automatic pattern detection from review findings and user corrections

052 GitHub stars

06Trend analysis to visualize the health and evolution of code patterns over time

사용 사례

01Tracking the effectiveness of technical debt reduction through trend analysis and frequency counting

02Standardizing code quality and architectural patterns across large microservice architectures

03Automating the maintenance of team-specific engineering runbooks and documentation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add datum-cloud/claude-code-plugins learning-engine

For use in Claude.ai and ChatGPT

주요 기능

01Multi-factor confidence scoring based on frequency, severity, and recency

02Cross-service pattern matching to identify systemic issues across repositories

03Automated runbook generation and promotion for high-confidence insights

04Automatic pattern detection from review findings and user corrections

052 GitHub stars

06Trend analysis to visualize the health and evolution of code patterns over time

사용 사례

01Tracking the effectiveness of technical debt reduction through trend analysis and frequency counting

02Standardizing code quality and architectural patterns across large microservice architectures

03Automating the maintenance of team-specific engineering runbooks and documentation