What makes this different from standard code search?

Unlike basic search, this skill performs iterative analysis across five different technical perspectives, tracing actual execution paths rather than just matching keywords.

How does it ensure the accuracy of its findings?

It adheres to a 'Zero Tolerance for Shallow Research' policy, which requires specific file paths and line numbers as evidence for every technical claim it makes.

When is the best time to activate the Wiki Researcher?

Use it when you need to understand 'how' a complex system works across many files, such as investigating architectural patterns or performing a deep technical audit.

Does it support visual diagrams?

Yes, the skill can generate Mermaid diagrams to visualize complex architecture and data flows discovered during its research iterations.

Deep Wiki Code Researcher

Name: Deep Wiki Code Researcher
Author: shouenlee

byshouenlee

0•

Learning & Documentation

Performs multi-turn, evidence-based architectural analysis to uncover deep implementation details within complex codebases.

The wiki_researcher skill is a sophisticated analysis engine designed for developers who need to understand complex, multi-file systems without relying on surface-level assumptions. By following a rigorous five-iteration process—covering architecture, data flow, dependencies, patterns, and synthesis—it traces actual code paths and cites specific file lines to ensure every technical claim is grounded in reality. It is an essential tool for onboarding to new projects, debugging intricate data flows, or conducting thorough technical debt audits where shallow analysis is insufficient.

Key Features

01Iterative Five-Lens Analysis covering architecture, data, integration, and patterns

02Automatic Mermaid diagram generation for visualizing system architecture

03Evidence-based path tracing with mandatory file and line-number citations

040 GitHub stars

05Zero-tolerance policy for shallow analysis and 'vibes-based' summaries

06Confidence-rated findings to distinguish between verified code and inference

Use Cases

01Tracing end-to-end data flow for complex state management or API lifecycle issues

02Performing a structural audit to identify technical debt and architectural anti-patterns

03Rapidly onboarding to a massive legacy codebase with undocumented dependencies

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add shouenlee/ghcp-dev-plugin wiki_researcher

For use in Claude.ai and ChatGPT

Key Features

01Iterative Five-Lens Analysis covering architecture, data, integration, and patterns

02Automatic Mermaid diagram generation for visualizing system architecture

03Evidence-based path tracing with mandatory file and line-number citations

040 GitHub stars

05Zero-tolerance policy for shallow analysis and 'vibes-based' summaries

06Confidence-rated findings to distinguish between verified code and inference

Use Cases

01Tracing end-to-end data flow for complex state management or API lifecycle issues

02Performing a structural audit to identify technical debt and architectural anti-patterns

03Rapidly onboarding to a massive legacy codebase with undocumented dependencies