REA
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Skills for AI coding agents.
Wenn du das gebaut hast, füge diesen Code in dein README ein. Das Badge verlinkt auf diese Seite.
[](https://mcp.popnx.com/en/skills/benjitaylor-skills)MCP-Server, die die KI für dieselbe Arbeit mit Diensten verbinden
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Codebase intelligence for TypeScript and JavaScript, exposed as structured MCP analysis tools.
MCP server for terminal commands, file operations, and process management
Gleiche Art Arbeit, sortiert wie die Skill-Liste
| Dieser skills | rea | andrej-karpathy-skills | graphify | MoneyPrinterTurbo | |
|---|---|---|---|---|---|
| Anbieter | benjitaylor | morluto | multica-ai | Graphify-Labs | harry0703 |
| Skills | 1 | 1 | 1 | 1 | 1 |
| GitHub-Sterne | 3.525 | 465.460 | 233.910 | 149.364 | 144.149 |
| Letzte Codeänderung | 5. Nov. 2026 | 17. Nov. 2026 | 20. Apr. 2026 | 17. Nov. 2026 | 17. Nov. 2026 |
| Lizenz | – | MIT | – | Apache-2.0 | MIT |
1 Skills (Anweisungssammlungen) von benjitaylor, die eine KI liest und befolgt · kein MCP-Server
MCP = die Hände
Verbindet die KI mit einem externen Dienst oder Daten, etwa Notion lesen oder eine Rechnung erstellen. Ein Programm oder Server muss laufen.
Skill = das Handbuch
Eine Textdatei, die der KI zeigt, wie eine Aufgabe geht. Sie verbindet sich mit nichts.
Manches überschneidet sich: Einige Skills starten Skripte mit Werkzeugen, einige MCP-Server liefern nur Wissen. Geh von der Aufgabe aus.
Cursor
Laut README des Anbieters und den Ordnern im Repo
GitHub account benjitaylor (person), 9 years old
★ 3.525 stars · code last changed 5. Nov. 2026
28 scripts that can run on a computer: assets/app-icons.ts, kit/act.ts, kit/brands/index.ts, kit/clock.ts and 24 more
Could not check every script (some are too large); read them first
No hooks or writes to machine settings found
Checked automatically from the repo files at commit 299140f (read only, nothing was run). Not a safety guarantee: read it and decide for yourself
Zuerst: Welche KI nutzt du?
Wähle deine App, dann zeigt diese Seite nur die Installation für diese App
This skill has no license, so we do not hand out a ZIP from our site. Download it from GitHub and arrange the folder yourself, or use "Let the AI install it" above, which is easier
Download from GitHub (whole repo) ↗
Unzip it, then zip only the stageshow folder again. The ZIP must have a stageshow folder with SKILL.md at the top level, no loose files
+Create skillUpload a skillpick stageshow.zip
SettingsCapabilitiesCode execution and file creation
Turn on Code execution and file creation first (newer menus: Create and edit files) · personal plans: Settings → Capabilities · Team/Enterprise: an admin turns it on under Organization settings → Plugins & skills · every plan, including Free
Upload once; it works in claude.ai and Claude Desktop, and in recent Claude Code signed in with a Claude account (not with an API key)
Cowork and cloud sessions use the skills turned on in your claude.ai account, not ~/.claude/skills on your computer (Claude Desktop scheduled tasks run locally and still load it)
Checked against official docs 12. Nov. 2026 · Source: support.claude.com, support.claude.com, code.claude.com
Kopiere diesen Satz und füge ihn in einen Claude Code-Chat ein
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Claude Code neu starten, dann /stageshow tippen oder eine passende Aufgabe stellen; Claude nutzt den Skill
Checked against official docs 12. Nov. 2026 · Source: code.claude.com
Ein Skill sind Anweisungen, denen die KI folgt. Lies die Dateien vor der Installation und installiere nur von vertrauenswürdigen Anbietern.
Schon installiert? Zuerst prüfen · Öffne das Terminal, gib diesen Befehl ein und drücke Enter
node -vMac: ⌘ + Leertaste drücken, Terminal eingeben, Enter
Windows: Start öffnen, Windows PowerShell eingeben, dann Öffnen
Noch nicht installiert: Öffne die offizielle Download-Seite, wähle die LTS-Version und den Klick-Installer: .pkg auf dem Mac · .msi unter Windows. Doppelklicken und bis zum Ende durchklicken. Kein curl-Befehl nötig
Download-Seite von Node.js öffnen ↗
Danach das Terminal schließen, neu öffnen und denselben Befehl erneut ausführen. Es muss eine Versionsnummer erscheinen: node -v
Einmal installieren, gilt für alles, was Node.js braucht
npx skills add benjitaylor/skills --skill stageshowAus dem README des Anbieters; lädt den Code des Anbieters herunter und führt ihn aus
Mac: Command+Leertaste und Terminal tippen · Windows: Der Befehl unten funktioniert nicht in PowerShell; nutze oben „KI installiert es“
Lädt das Repo in einen temporären Ordner und kopiert nur Ordner mit einer SKILL.md
git clone --depth 1 https://github.com/benjitaylor/skills.git /tmp/popnx-skill-benjitaylor-skills-299140f && mkdir -p ~/.claude/skills && cp -R '/tmp/popnx-skill-benjitaylor-skills-299140f/stageshow' ~/.claude/skills/stageshowSelbst installierte Skills funktionieren in der ChatGPT-Desktop-App, in Codex CLI und in der Codex-Erweiterung für die IDE · ChatGPT im Web und auf dem Handy kann selbst installierte Skills nicht nutzen, nur Skills, die mit einem Plugin aus dem Verzeichnis kommen
Die ChatGPT-Desktop-App enthält Codex, und Codex liest Skills aus dem Ordner ~/.agents/skills
Öffne Codex in der ChatGPT-Desktop-App und füge diesen Satz in den Chat ein
The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Lege den Ordner stageshow unter ~/.agents/skills/stageshow ab
~ ist dein Home-Ordner ($HOME)
Ein Skill sind Anweisungen, denen die KI folgt. Lies die Dateien vor der Installation und installiere nur von vertrauenswürdigen Anbietern.
Mac: Command+Leertaste und Terminal tippen · Windows: Der Befehl unten funktioniert nicht in PowerShell; nutze oben „KI installiert es“
Lädt das Repo in einen temporären Ordner und kopiert nur Ordner mit einer SKILL.md
git clone --depth 1 https://github.com/benjitaylor/skills.git /tmp/popnx-skill-benjitaylor-skills-299140f && mkdir -p ~/.agents/skills && cp -R '/tmp/popnx-skill-benjitaylor-skills-299140f/stageshow' ~/.agents/skills/stageshowChecked against official docs 15. Nov. 2026 · Source: learn.chatgpt.com
Selbst installierte Skills funktionieren hier nicht, nur Skills, die mit einem Plugin aus dem Verzeichnis kommen
Um diesen Skill mit ChatGPT zu nutzen, verwende die ChatGPT-Desktop-App wie oben beschrieben
Checked against official docs 15. Nov. 2026 · Source: learn.chatgpt.com
Kopiere diesen Satz und füge ihn in einen Codex-Chat ein · Funktioniert in Codex in der ChatGPT-Desktop-App, in Codex CLI und in der IDE-Erweiterung
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Codex erkennt neu installierte Skills selbst; wenn einer nicht erscheint, starte Codex neu · So nutzt du ihn: In ChatGPT @ tippen und den Skill wählen · in Codex CLI oder der IDE-Erweiterung /skills ausführen oder $ und den Skill-Namen tippen
Checked against official docs 15. Nov. 2026 · Source: learn.chatgpt.com
Ein Skill sind Anweisungen, denen die KI folgt. Lies die Dateien vor der Installation und installiere nur von vertrauenswürdigen Anbietern.
Mac: Command+Leertaste und Terminal tippen · Windows: Der Befehl unten funktioniert nicht in PowerShell; nutze oben „KI installiert es“
Lädt das Repo in einen temporären Ordner und kopiert nur Ordner mit einer SKILL.md
git clone --depth 1 https://github.com/benjitaylor/skills.git /tmp/popnx-skill-benjitaylor-skills-299140f && mkdir -p ~/.agents/skills && cp -R '/tmp/popnx-skill-benjitaylor-skills-299140f/stageshow' ~/.agents/skills/stageshowKopiere diesen Satz und füge ihn in einen Cursor-Chat ein
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Im Agent-Chat von Cursor / tippen und stageshow wählen
Checked against official docs 12. Nov. 2026 · Source: cursor.com
Ein Skill sind Anweisungen, denen die KI folgt. Lies die Dateien vor der Installation und installiere nur von vertrauenswürdigen Anbietern.
Mac: Command+Leertaste und Terminal tippen · Windows: Der Befehl unten funktioniert nicht in PowerShell; nutze oben „KI installiert es“
Lädt das Repo in einen temporären Ordner und kopiert nur Ordner mit einer SKILL.md
git clone --depth 1 https://github.com/benjitaylor/skills.git /tmp/popnx-skill-benjitaylor-skills-299140f && mkdir -p ~/.cursor/skills && cp -R '/tmp/popnx-skill-benjitaylor-skills-299140f/stageshow' ~/.cursor/skills/stageshowFür andere Apps, die Skills lesen. Der Satz enthält die Pfade für Claude Code, Codex und Cursor, damit die KI den für deine App wählt · Copy this sentence into the app’s chat
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Checked against official docs 12. Nov. 2026 · Source: code.claude.com, learn.chatgpt.com, cursor.com
Andere Apps mit Agent Skills (etwa Gemini CLI oder GitHub Copilot) nutzen andere Ordner. Sieh in deren Anleitung nach und kopiere den Ordner mit der SKILL.md dorthin.
Skill-Ordner am 11. Nov. 2026 mit den offiziellen Anleitungen von Claude Code, Codex und Cursor geprüft
✓ Befehle geprüft am 18. Nov. 2026
Der folgende Text stammt aus dem Repo des Anbieters, nicht von PopNx MCP
Die ersten 8.000 Zeichen · Ganze Datei auf GitHub öffnen
---
name: stageshow
description: Build faux product UI that looks like a real product, as a still illustration or an animated demo with a cursor. Browser and app windows, dashboards, phones, notifications, terminals, bento grids and cursor walkthroughs, framed, masked and cropped the way a marketing page, feature card or docs hero needs them. Use whenever someone wants a fake or mocked UI, a product demo or walkthrough animation, a hero or feature-card illustration of an app, a product shot in the style of a polished marketing site, or an animated cursor doing things in an interface, even if they never say "faux UI". React and plain CSS.
metadata:
version: 1.1.1
---
# Stageshow
Miniature product UI: a believable interface, framed in its space, as one still or a short animation in which one thing happens. The viewer recognizes the screen and sees the thing happen.
## The one idea
**It reads as a real product, simplified to what the viewer needs, framed with intent, and every motion has a cause.**
Five ways it fails:
- **Too much.** Real text in every row, numbers everywhere, everything packed. The more detail, the more it looks generated.
- **Wireframe.** Two gray blobs and a button. Built too small, or stripped of what made it recognizable.
- **Boxed.** A whole window shrunk into a card with even margins: a screenshot pasted in, not a product shot.
- **Flat.** Every object the same weight in the same bordered card. The eye has nowhere to land.
- **Bland.** A gray window with a cursor, whatever the brief said. The brief's look and energy (a blueprint, glass over color, a Dynamic Island, a mascot) are the point; take what it asks for and build it to the rules, as `examples/hero-demo` does with its typed popup and terminal satellite (a satellite is a smaller second object, a receipt, phone or detail card, straddling the hero's edge).
The fix: a full, structured screen in quiet grays, real words only on what the story is about, placeholder lines for the rest, room to breathe, a crop that puts the eye where the story is. Simplify the words, not the context. An empty corner, a void under a list, rows spread to fill a pane, a window floating on a plate (a backdrop or tinted field behind it): unfinished, not calm.
These are authored illustrations, not browser automation: `a.click()` draws the cursor pressing, `a.set()` changes the scene.
## How hard to press
Build it the way a product designer builds a launch page they will be judged on. The gold examples in `examples/` are the bar, not a reference: if yours is the emptier, flatter or smaller one beside the closest gold, that is the first fix. Then look at the frame the way the most exacting designer you know would, the one who screenshots the one wrong thing and sends it back with "what is this?": a dot off its avatar, a unit on its own line, a grabber under the island, copy touching a card. If you can picture that screenshot, you have found the next fix.
## Workflow
Most of the quality is decided before any code. Write the plan first: where it sits, the focal object, the action, the payoff, what can be left out.
1. **Medium and place.** Still or animation (loop, once in view, or hover). The page, its ground, its width. Design at the size it will be seen, about 720 × 340 for a hero beside copy (`hero-demo`), 900 to 1100 wide for a full-width hero (`linear`), 400 × 300 for a card, shown at 1:1; never under about 320 wide for a demo, though a bento tile or a satellite card may be smaller. Every demo that can sit in a phone column ships a `compact` design at the column's width (358), a redesign not a squeeze (`references/framing.md`, Small formats and the page). A grid of tiles is one set: plan it with `references/bento.md`.
2. **Look.** With the product's code or site at hand, map its tokens to the kit's (`references/looks.md`); if the developer asks for their own components, use them inside the stage instead of the kit's small parts. Otherwise a theme (light, dark, blueprint, glass), a palette (accent, mono) and a fidelity: `minimal` is all bars, `standard` is words where they carry meaning and bars elsewhere, `realistic` is real content everywhere. Default light, accent, standard; go realistic at hero scale, for a named real product, or a still whose subject is a list or page. Write filler once with `Text filler` and the fidelity decides. The product's own icons, its real chrome, the content that makes it recognizable. A brief with a style (kawaii, paper, a poster color) gets that style in full, still built to these rules.
3. **The story.** Its shape (`references/motion.md`, Story): how it feels to use, a flow, or what the system does. Three to five beats and a payoff, 10 to 16 seconds. No cursor move you can't explain. Data never rewinds in view without a cause. Between a richer story and a calmer frame, choose the frame.
4. **Start from an example.** Find the closest gold example (the Gold table below) and film it (`node scripts/film.mjs "<preview>/?capture=<gold>" --select '#example' --still --out gold.png`; `npm run dev` serves the preview). Copy its folder into your project as the starting point; don't rewrite it from a glance. The quickest loop: copy it to `examples/<yours>/` here, where the preview picks it up and `?capture=<yours>` films it, then move it into the host when it passes and fix its kit import path. Keep its structure: the stage size, the crop and bleed, row and type sizes, spacing, the script's beats, the kit parts. Now read `references/craft.md` in full: it is short, and it holds the still-frame rules the gold can't show you, the cases your gold doesn't happen to contain. Then replace its look and content with the brief's product: the tokens from step 2, the product's own icons, chrome and words. A copy that only changes the name is not done: it should read as the brief's product, not the gold's. If no gold tells your story, choose in two parts: the layout first (pick it from the table in `references/framing.md`; each layout names its example), and copy that example even though its story differs; then borrow the motion for your beats from the example whose story is closest, and build what neither has in the same style. Then frame it: crop to the panel the story happens in, judged in the final visible crop (`references/framing.md` has the rules and the layout table). The short form: one gutter, a full one or well past; only filler bleeds; what opens stays in; windows keep their shape; the page is the ground, so no plate under a hero on a white or dark page and no window whole with air on four sides; `unframed` only when the developer asks; no dead corners. Film one still at both widths beside the gold's; if yours is emptier, flatter or more boxed, fix the frame before any motion.
5. **The screen.** Count the real strings, usually 3 to 7 (`references/craft.md`, Restraint): a budget for words, not elements. Build with the kit's tokens, `Bar`, `Text` and `Row`. Fit every control's text before it moves. Mark what the cursor touches with `data-target` and `data-cursor`; measure with `rect()`, never hard-code.
6. **Choreograph.** The `Act` API. Keep what appears mounted and toggle it so exits animate (`Bubble` and `Fly` hand off explicitly). Animate the relationship, not just the object: a new row carries the rows below it, a surface that resizes never stretches its text. Beats overlap, groups stagger, the payoff holds, the loop ends where it began or resets with a visible cause.
7. **Rehearse.** Set `poster` first: most people see only that frame (a thumbnail, reduced motion, a screenshot), so it is the beat that tells the story on its own, never one caught mid-change, with the result left on screen rather than tidied away (`references/motion.md`, Story). Then film the loop, not just a still, and always with `--gate`: it is the same contact sheet plus the verdict, so it costs nothing extra. Film the contact sheet at the shown width and at 390, and step through every press, mFragen, Tipps teilen oder melden, dass etwas nicht geht. Sortiert nach „hilfreich“.
Noch keine Beiträge. Stell die erste Frage oder teile einen Tipp.
Beiträge hier stammen von Nutzern, nicht vom PopNx-Team.