how

Explains codebase architecture and subsystem flows using parallel explorer and critic agents.

136|8|Updated May 9, 2026
One-click install
npx skills add https://github.com/Sma1lboy/rove --skill how-sma1lboy
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: how
Source: https://github.com/Sma1lboy/rove/tree/main/.agents/skills/pstack/skills/how
Command: npx skills add https://github.com/Sma1lboy/rove --skill how-sma1lboy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Understanding how an unfamiliar subsystem works requires hours of manual code reading, and architectural questions like "where should this live" or "is this the right layer" lack a structured way to get rigorous answers. ## Core Features & Use Cases - Explain Mode: Answers "how does X work" questions by spawning parallel read-only explorer agents that trace code paths, then synthesizing their findings into a senior-engineer-level architectural explanation with file references, flow walkthroughs, and gotchas. - Critique Mode: After explaining a subsystem, spawns multiple independent model critics that evaluate the architecture against a rubric covering abstraction fit, data model, boundary discipline, and evolution readiness, with findings triaged by a lead judgment pass. - Use Case: Before modifying a rate limiter you have never touched, ask "how does the rate limiter work?" to get a mental model of its data flow, key types, and sharp edges, then optionally request an architectural critique before proposing changes. ## Quick Start Ask how the authentication flow works in this repository to receive a structured architectural explanation.

Frequently Asked Questions about how

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

FAQPage Schema
How do I understand how a subsystem works before changing it?▼

Ask a "how does X work" question and the skill explores the codebase, traces the flow from entry point to output, and returns a structured explanation covering key concepts, step-by-step flow, file locations, and gotchas.

How does the skill handle simple versus complex questions?▼

Simple questions about a single module are explored and explained in one pass by a single agent. Complex questions spanning multiple files trigger 2-4 parallel explorer agents whose findings are reconciled by a synthesizing explainer.

Can it critique architecture instead of just explaining it?▼

Yes. Critique mode first produces the full explanation, then spawns one critic per configured model to independently find architectural issues using a rubric covering abstraction fit, data model, boundaries, and evolution readiness.

Does the exploration modify any code in the repository?▼

No. All explorer, explainer, and critic agents run in read-only mode. They use Glob, Grep, and Read to inspect code and return findings without writing or changing any files.

What kinds of questions is this skill not suited for?▼

It targets "how does X work" and placement or layering questions. Motivation questions like "why was this built" belong to a separate why skill, and line-level bug review is explicitly out of scope for its critics.