What are the three tiers of documentation involved?

Tier 1: Architecture Patterns (Conceptual/Rules), Tier 2: Reference Implementations (Concrete file examples), and Tier 3: Code Snippets (Ready-to-use boilerplate).

When should I update my pattern documentation?

You should update the documentation whenever a pattern is refined, a new canonical example is created, or when the 'last_verified' metadata indicates the documentation may be stale.

Is this system compatible with any programming language?

Yes, the pattern extraction methodology is language-agnostic and can be applied to any tech stack, including frontend, backend, and infrastructure code.

How does this skill save tokens during development?

It provides small, curated documentation files (~2k tokens) that explain patterns, preventing the AI from needing to read large source files (10k+ tokens) repeatedly to understand how to implement features.

Code Pattern Documentation System

Name: Code Pattern Documentation System
Author: violetio

byvioletio

•

Productivity & Workflow

Optimizes AI token usage and development consistency by implementing a three-tier code pattern documentation workflow.

The Pattern Extraction skill provides a standardized framework for documenting codebase architectures, reference implementations, and reusable snippets. By categorizing knowledge into conceptual rules, concrete examples, and boilerplate templates, it allows AI agents and developers to understand and replicate complex patterns without expensive, repetitive exploration of the entire repository. This system is particularly useful for maintaining consistency across large-scale projects and onboarding AI agents to specific architectural styles while significantly reducing token consumption.

Key Features

011 GitHub stars

02Standardized metadata for pattern verification and staleness detection

03Repeatable workflow for extracting and updating patterns from live codebases

04Three-tier documentation hierarchy (Architecture, References, Snippets)

05Token-efficient knowledge transfer for AI agents and developers

06Clear separation between conceptual rules and copy-pastable boilerplate

Use Cases

01Onboarding AI agents to large, complex legacy codebases without exhaustive scanning

02Reducing token costs during feature implementation by providing concise context

03Enforcing architectural consistency across multiple microservices or frontend applications

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add violetio/violet-ai-plugins pattern-extraction

For use in Claude.ai and ChatGPT

Download Skill