What documentation formats does this skill support?

The skill generates three specific formats: Pattern Guides (for best practices), API Manuals (for library references), and Architecture Decision Records (ADRs) following the Nygard format.

How does it handle different programming languages?

It includes a Phase 0 Stack Detection step that identifies files like package.json or *.csproj to adapt its research queries and documentation style to your specific language environment.

What tools does the researcher use for its data?

It utilizes MCP Ref for documentation searching, Context7 for library-specific details, and WebSearch to verify the latest stable versions and security standards like OWASP.

Why are there no code examples in the generated documentation?

The skill follows a strict 'No-Code' policy to maximize maintainability. It uses tables, ASCII diagrams, and links to official documentation to prevent the reference material from becoming outdated as syntax changes.

Best Practices Researcher

Name: Best Practices Researcher
Author: levnikolaevich

bylevnikolaevich

•

Learning & Documentation

Researches industry standards and generates professional project documentation including pattern guides, API manuals, and Architecture Decision Records (ADRs).

About

The Best Practices Researcher is a sophisticated Claude Code skill designed to automate the creation of high-quality, professional technical documentation. By leveraging specialized MCP tools like Ref and Context7, it researches the latest industry standards, security patterns, and library versions for 2025 and beyond. The skill intelligently detects your project's technology stack—supporting .NET, Node.js, Python, Go, Rust, and Java—to produce contextually relevant Guides, API Manuals, and Nygard-style ADRs. It strictly enforces a maintenance-friendly 'no-code' documentation policy, prioritizing structured tables, ASCII diagrams, and official links to ensure your project's reference materials remain evergreen and easy to read.

Key Features

Integrated Q&A dialog for capturing architectural decision rationale.
Generates three document types: Pattern Guides, API Manuals, and ADRs.
Strict documentation standards utilizing tables and ASCII-schemes over code snippets.
Automatic stack detection for .NET, Node.js, Python, Go, Rust, and Java.
38 GitHub stars
Automated research via MCP Ref and Context7 for current industry standards.

Use Cases

Establishing standardized pattern guides with Do/Don't/When tables for development teams.
Documenting complex architecture decisions through structured ADR workflows.
Generating up-to-date API manuals for third-party libraries using official documentation links.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add levnikolaevich/claude-code-skills ln-002-best-practices-researcher

For use in Claude.ai and ChatGPT

Download Skill

GitHub

About

Key Features

Integrated Q&A dialog for capturing architectural decision rationale.
Generates three document types: Pattern Guides, API Manuals, and ADRs.
Strict documentation standards utilizing tables and ASCII-schemes over code snippets.
Automatic stack detection for .NET, Node.js, Python, Go, Rust, and Java.
38 GitHub stars
Automated research via MCP Ref and Context7 for current industry standards.

Use Cases

Establishing standardized pattern guides with Do/Don't/When tables for development teams.
Documenting complex architecture decisions through structured ADR workflows.
Generating up-to-date API manuals for third-party libraries using official documentation links.