investigate

Explore codebases with parallel sub-agents using glob and grep operations.

Updated Mar 12, 2026

Search Tags:#investigation #codebase analysis #code exploration #dependency mapping #sub-agents #parallel research

Ready to run

npx skills add https://github.com/sogadaiki/claude-code-skills --skill investigate-sogadaiki

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/sogadaiki/claude-code-skills/tree/main/skills/investigate
Command: npx skills add https://github.com/sogadaiki/claude-code-skills --skill investigate-sogadaiki

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill allows for deep code exploration and investigation without cluttering the main AI context, enabling parallel research and analysis.

Core Features & Use Cases

Parallel Exploration: Utilizes sub-agents to concurrently search and analyze codebases.
Context Preservation: Keeps the main AI context clean and focused by offloading research tasks.
Use Case: Investigate how a specific feature like "user authentication" is implemented across multiple projects or modules, identifying all relevant files and dependencies.

Quick Start

Use the investigate skill to find all implementations of the Stripe payment gateway.

Frequently Asked Questions about investigate

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

FAQPage Schema

How do I explore a large codebase for dependency analysis without cluttering the main context?▼

You can conduct codebase exploration using parallel sub-agents. This approach offloads research tasks like dependency analysis and error tracing, keeping the main AI context clean and focused while concurrently searching and analyzing code structure.

What is the best way to trace how a specific feature is implemented across multiple modules?▼

Tracing feature implementations across modules is done by deploying dedicated sub-agents for parallel research. These agents execute glob and grep operations concurrently to identify all relevant files and map dependencies without overwhelming the primary context window.

How do I map dependencies and analyze code structure across different projects?▼

Dependency mapping and cross-project research are performed by parallel sub-agents. They apply pattern analysis to investigate code structure and trace dependencies across multiple projects simultaneously, yielding comprehensive exploration results.

Can I investigate cross-project code structure and error tracing in parallel?▼

Yes, parallel exploration allows you to investigate code structure and trace errors across projects concurrently. Sub-agents handle the glob and grep operations independently, enabling deep dives into multiple projects without serializing the research process.

Does this codebase exploration method work for deep dives into complex code structure?▼

Yes, this method is designed for deep code exploration and investigation. Sub-agents execute pattern analysis and dedicated search operations specifically to handle complex code structure scenarios and deep dependency mapping.

Why does parallel code exploration keep the main AI context clean during research?▼

Parallel code exploration preserves context by offloading research tasks to sub-agents. These dedicated agents handle the heavy lifting of pattern analysis and file searching independently, preventing the main context from being cluttered with intermediate search data.