research-codebase

Document codebase architecture and behavior using parallel sub-agents.

2|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/fredrick84823/fstack --skill research-codebase-fredrick84823
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-codebase
Source: https://github.com/fredrick84823/fstack/tree/main/skills/research-codebase
Command: npx skills add https://github.com/fredrick84823/fstack --skill research-codebase-fredrick84823

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Document codebases as-is by coordinating parallel sub-agents to map architecture, trace dependencies, and answer questions about existing systems.

Core Features & Use Cases

  • Parallel discovery and mapping using codebase-locator, codebase-analyzer, and codebase-pattern-finder to surface structure and behavior without changing the code.
  • Thoughts integration: auto-detects thoughts/ directory (if present) to incorporate historical context and syncing decisions.
  • Use cases include onboarding new team members, architecture audits, and answering system questions with structured reports.

Quick Start

Provide a research question and let parallel sub-agents map the codebase and generate a structured report.

Frequently Asked Questions about research-codebase

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

FAQPage Schema
How do I document codebase architecture and map module relationships?

Document codebase architecture by deploying parallel sub-agents that analyze components and trace dependencies to generate structured reports explaining module relationships and system behavior. It automatically detects a thoughts directory to incorporate historical context if present.

What is the best way to analyze a codebase for onboarding and architecture audits?

Analyzing a codebase for onboarding and architecture audits is best handled by parallel sub-agents that discover structural patterns and map data flows without modifying the code. They produce structured reports detailing interactions and historical decisions.

Can I research codebase behavior across different programming languages?

Researching codebase behavior applies to diverse codebases across multiple languages and sizes. The parallel sub-agents adapt to map architecture, trace dependencies, and answer questions about data flows regardless of the underlying language.

How do I start tracing data flows and dependencies in an existing codebase?

Start tracing data flows by providing a research question. Parallel sub-agents automatically map codebase components, analyze interactions, find structural patterns, and sync historical decisions to deliver a comprehensive answer in a structured report.

Does codebase analysis work with a thoughts directory for historical context?

Codebase analysis auto-detects a thoughts directory if it exists and integrates it to incorporate historical context and syncing decisions. This provides deeper insight into past architecture choices when mapping components and interactions.