how

Explain how a codebase subsystem works and critique its architecture with parallel read-only agents.

20|2|Updated May 16, 2026
One-click install
npx skills add https://github.com/reddb-io/red-skills --skill how-reddb-io
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: how
Source: https://github.com/reddb-io/red-skills/tree/main/plugins/dev/skills/engineering/how
Command: npx skills add https://github.com/reddb-io/red-skills --skill how-reddb-io

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Engineers often need to understand an unfamiliar subsystem before changing it, but tracing code paths across files and packages is slow and error-prone. This Skill answers "how does X work" questions by exploring the codebase and producing a senior-engineer-level explanation, and can optionally critique the architecture with an independent panel. ## Core Features & Use Cases - Explain mode: Decomposes complex questions into parallel exploration angles, synthesizes findings into a structured explanation with overview, key concepts, flow, file map, and gotchas. - Critique mode: After explaining, spawns 3-4 diverse critics using a rubric covering abstraction fit, data model, boundaries, evolution readiness, complexity, and consistency, then sorts findings into Act on / Consider / Noted / Dismissed buckets. - Use Case: Before modifying a rate limiter, ask how it works; the Skill traces the data model, request path, and configuration in parallel, then delivers a walkthrough you can onboard from or critique. ## Quick Start Ask the assistant to explain how a specific subsystem in this repository works, for example how the authentication flow handles token refresh.

Frequently Asked Questions about how

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

FAQPage Schema
How do I understand how an unfamiliar codebase subsystem works?▼

Ask a focused "how does X work" question and the Skill traces the code from entry point to output using read-only explorer agents. Simple questions get a single-pass explanation; complex subsystems are decomposed into parallel exploration angles and synthesized into one walkthrough.

How to review the architecture of a module before refactoring it?▼

Request a critique alongside the explanation. The Skill first explains the subsystem, then spawns 3-4 independent critics with distinct rubric emphases, and a lead judgment sorts findings into Act on, Consider, Noted, and Dismissed buckets with concrete code evidence.

Can this answer where new code should live in a project?▼

Yes, placement and layering questions like "where should this live" or "is this the right layer" are in scope. The explanation maps boundaries and file locations so you can judge the correct layer for new code.

When should I use code explanation versus debugging or research tools?▼

Use this Skill for understanding how existing code in this repository works. Bugs and regressions belong to a diagnose workflow, external official documentation belongs to research, and repo-wide refactoring hunts belong to a dedicated architecture-improvement skill.

Does the architecture critique modify any code?▼

No, all explorer and critic agents are strictly read-only. They read files, trace flows, and report findings with evidence; they never edit the codebase, so it is safe to run on production repositories.