show-me Skill (humanlayer) for Claude: What It Does and How to Install It
Makes Claude explain what you are discussing with a small diagram, tree, diff or HTML page instead of long prose.
- Installs
- 34.2K
- on skills.sh
- Stars
- 5.5K
- on GitHub
- Files
- 1
- besides SKILL.md
- Updated
- Nov 2026
- last repo commit
show-me makes Claude pick the smallest visual that makes the point. That can be pseudocode, a call tree, a component tree, a file tree, a Mermaid diagram or a diff, and for dense topics one focused HTML page that it opens for you.
You run it yourself with /show-me. The SKILL.md sets disable-model-invocation: true and the agents file turns implicit invocation off, so it is not meant to fire on its own. It bundles no scripts.
Good for
- Developers who want a quick picture of a refactor before approving it
- Reviewers trying to follow control flow or file layout
- Anyone who learns faster from a diagram than a paragraph
Install show-me
Claude Code
- 1
Copy this prompt.
Install the show-me skill in Claude Code. Run this command: npx skills add https://github.com/humanlayer/skills --skill show-me Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Requires Node.js.
From its listing on skills.sh. Source.
What's in the skill
When it loads
Help the user understand the current topic visually with concise diagrams, code-shape sketches, and focused HTML artifacts.
Sections in SKILL.md
- guidance
Key facts
- Repository
- humanlayer/skills
- Installs on skills.sh
- 34,187 as of 2026-11-02
- License
- MIT
- Last commit to the repo
- 2026-11-01
- What is inside
- 2 files including SKILL.md: agents/openai.yaml
- How to run
- /show-me (README)
- skills.sh audits
- Gen Agent Trust Hub pass, Socket pass, Snyk pass
What it includes
Show logic or an algorithm
A short text block of the steps, for example on(save) with a cache check.
Read ·
PseudocodeShow runtime control flow
A nested list such as submitForm, createSession, persistPrompt.
Read ·
Call treeShow UI structure with state and module boundaries
A TSX-style outline with file paths where they matter.
Read ·
Component treeShow file responsibility or a broad refactor
A shallow tree with one-line comments for each folder.
Read ·
File treeShow component interaction or data flow
Sequence or flow diagrams for how parts talk to each other.
Read ·
Mermaid diagramShow what changes when the shape already exists
Four diff styles for component trees, file layout, call trees and state or control flow.
Read ·
Diff viewsShow the whole block when most of it is new
For when omitted context would hide order or ownership, or you need a copyable target shape.
Read ·
Full code blockOne diagram, infographic or short slide deck
For a visual UI or a concept too dense for Mermaid. Claude writes show-me-{description}.html, matches your product's colors and type, and opens it with the open command.
Try: “/show-me how the checkout flow moves data between the UI and the API”
Runs code ·
Focused HTML page
From the official docs
Permissions
Low risk
When it makes an HTML page, Claude writes a file and opens it with the open command. Approve the command only for a path you recognize.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
It writes an HTML file and opens it
The SKILL.md example runs the open command on the new file. Check the file path before you approve it. github.com
Technical details
| Turning on read-only | No read-only switch is documented in SKILL.md. |
|---|---|
| Scopes |
|
Questions
What does show-me do?
It makes Claude explain the current topic visually and briefly, choosing one or a few formats: pseudocode, trees, Mermaid, diffs or an HTML page.
How do I install show-me?
Use the install box at the top of this page. The README's own command installs this skill alone from the humanlayer/skills repo.
How do I run it?
Type /show-me in Claude Code, per the README. The skill is set not to trigger by itself, so you start it when you want a picture.
Is show-me safe to use?
It bundles no scripts. For the HTML option, Claude writes an HTML file and runs the open command on it, so approve that command only for a path you recognize.
Does it work in Codex?
The skill folder includes agents/openai.yaml with implicit invocation turned off. The README documents the slash command only for Claude Code, so check how your Codex setup invokes skills.
Which other skills are in the repo?
The README lists rpi, visual-pr, improve-claude-md, narrow-react-prop-types, build-iterated-agentic-loop and design-control-loop. They are separate skills.
Information
updated Nov 2, 2026
- Repository
- humanlayer/skills
- Folder
plugins/show-me/skills/show-me
Sources
Docs, issue trackers and discussions this page is built from.
Show all 3Hide
- show-me SKILL.md · github.com
- humanlayer/skills README · github.com
- agents/openai.yaml · github.com