investigate

Investigate codebase questions using llmdoc documentation and report findings.

459|49|Updated Oct 10, 2025
One-click install
npx skills add https://github.com/TokenRollAI/cc-plugin --skill investigate-tokenrollai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate
Source: https://github.com/TokenRollAI/cc-plugin/tree/main/skills/investigate
Command: npx skills add https://github.com/TokenRollAI/cc-plugin --skill investigate-tokenrollai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Quickly answer user questions about codebase structure and behavior by consulting llmdoc-driven documentation and returning findings directly in conversation.

Core Features & Use Cases

  • Documentation-first exploration: read llmdoc index, overview, architecture, guides, and references before inspecting code.
  • Fast codebase insights: produce concise findings without persisting files.
  • Report-ready outputs: deliver structured, actionable conclusions for product and engineering decisions.

Quick Start

Ask for an explanation of a module by summarizing documented behavior from llmdoc.

Frequently Asked Questions about investigate

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

FAQPage Schema
How do I investigate a codebase to understand how a specific module works?

To investigate a codebase, consult llmdoc-first documentation by reading the index, overview, architecture, and guides before inspecting code. This approach uses Glob and Grep to gather information and returns concise findings directly without persisting files.

What is the best way to explore a codebase structure across multiple repositories?

Exploring codebase structure across repositories is best handled by querying llmdoc-driven documentation first. This method reads documented references and uses Glob, Grep, and Read to produce structured, actionable findings for engineering decisions.

Can I get codebase insights without generating or persisting extra files?

Yes, you can get codebase insights without persisting files. The investigation process reads llmdoc documentation and source code directly, delivering report-ready outputs and concise findings in conversation without creating any new files.

Does codebase investigation work with existing llmdoc references?

Yes, codebase investigation works with llmdoc references by design. It prioritizes reading the llmdoc index, overview, architecture, and guides to answer questions about codebase behavior before falling back to Glob and Grep on source code.

What are the limitations of a documentation-first codebase investigation approach?

A documentation-first codebase investigation relies on the accuracy and completeness of existing llmdoc files. If documentation is outdated or missing, the tool must fall back to searching code with Glob and Grep, which may yield less structured findings.