Skip to content
Agentmart
Submit a tool

how

Explain subsystem architecture, runtime flow, placement, ownership, or layering before changing code.

Skill by matthewyjiang

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/matthewyjiang-how

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Explain subsystem architecture, runtime flow, placement, ownership, or layering before changing code. Use for how-does-this-work questions and grounding a nontrivial design.

Use cases: design

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/how

What's inside

Skill name
how
Description
Explain subsystem architecture, runtime flow, placement, ownership, or layering before changing code. Use for how-does-this-work questions and grounding a nontrivial design.
License
MIT
Instructions
21 lines in SKILL.md · view on GitHub

Files (3)

  • SKILL.md
  • explainer-prompt.md · references/
  • explorer-prompt.md · references/

SKILL.md (first lines)

# How

Read [the workflow contract](../../references/workflow-contract.md).
Build a senior engineer's working mental model, not annotated source.

1. State the question and boundaries. For a known symbol or narrow module, trace
   it directly; do not delegate a trivial lookup.
2. For a broad unfamiliar subsystem, identify independent slices worth exploring.
   Batch `agent` calls to the available `explorer` with self-contained questions
   built from `references/explorer-prompt.md`. No model-per-call selection. If that
   ID is unavailable, inspect `rho` action `agents`, or investigate directly.
3. Await all required findings. Follow real entry points, types, ownership, state,
   calls and errors; cite file/line evidence and resolve contradictions in source.
4. Synthesize using `references/explainer-prompt.md`; usually do this in the parent,
   not another delegate just to rephrase. Keep unknowns visible.

Return only applicable sections: Overview, Key concepts, How it works, Where things
live, Gotchas. A small Mermaid diagram helps when control flow or relationships
are the point. Distinguish mechanism from rationale; use `why` for motivation.

From matthewyjiang/rho-pstack, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 17, 2026 (12 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from matthewyjiang

Cite this page

how skill. Agent Mart. https://agentmart.dev/skills/matthewyjiang-how (last updated 2026-10-24)