research_codebase

Coordinate parallel sub-agents to research codebases and produce a final document.

20|6|Updated Jan 16, 2026
One-click install
npx skills add https://github.com/adrielp/ai-engineering-harness --skill research-codebase-adrielp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research_codebase
Source: https://github.com/adrielp/ai-engineering-harness/tree/main/gemini/skills/research_codebase
Command: npx skills add https://github.com/adrielp/ai-engineering-harness --skill research-codebase-adrielp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables comprehensive codebase research by delegating tasks to parallel sub-agents and synthesizing findings using Gemini CLI tools.

Core Features & Use Cases

  • Spawns parallel sub-agents to read, analyze, and synthesize findings across the codebase.
  • Uses Gemini CLI tools to coordinate research, fetch relevant files, and compile results into a coherent report.
  • Use Case: when you need to quickly understand a large codebase, identify hotspots, dependencies, and architecture decisions.

Quick Start

Ask the Skill to research a specific question about the codebase and specify scope by module or file.

Frequently Asked Questions about research_codebase

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

FAQPage Schema
How do I analyze a large monorepo to understand its architecture and dependencies?

To analyze a large monorepo, this Skill coordinates parallel sub-agents that read and synthesize findings across the codebase, identifying architectural hotspots and dependencies to produce a coherent research report.

What is the best way to automate codebase research for multi-module projects?

Automating codebase research for multi-module projects is best handled by spawning parallel sub-agents to analyze files and synthesize findings, leveraging Gemini CLI tools to compile results into a final research document.

Can I use parallel agents to identify code hotspots and architectural insights?

Yes, you can use parallel agents to identify code hotspots and architectural insights by delegating analysis tasks across the codebase, fetching relevant files, and synthesizing the data into a comprehensive report.

How do I coordinate research across a large codebase using Gemini CLI?

You coordinate research across a large codebase using Gemini CLI by applying parallel task orchestration, file discovery, and analysis patterns to synthesize findings from multiple sub-agents into a final document.

Does this codebase analysis approach work for large scale projects?

Yes, this codebase analysis approach works for large scale projects, monorepos, and multi-module projects by distributing the research workload across parallel sub-agents to handle extensive file discovery and synthesis.