← Search

Taste Skill — Mobile Image Generation
Design 🧪 Install ran in our test sandbox · Oct 19, 2026 📌 Pinned version: commit beb16b0
Install test only, on the date shown. This does not guarantee the code is safe.
⚙ Needs: none for installation — no account, API key…

Taste Skill — Mobile Image Generation

An art-direction skill that turns Muse into a mobile product design director: describe an app screen or flow (onboarding, dashboard, chat, checkout…) and it generates premium, app-native screen concepts with coherent style, controlled palettes, and believable phone framing.

At a glance
What
An art-direction skill that turns Muse into a mobile product design director: describe an app screen or flow (onboarding, dashboard, chat, checkout…) and it generates premium, app-native screen concepts with coherent style, controlled palettes, and believable phone framing.
Cost
Free
Needs
none for installation — no account, API key, or paid plan required, and this skill is free. (Using the skill later requires a host agent with image-generation capability, for example ChatGPT Images, Codex image mode, or any agent that can generate images; installing the skill itself requires none of that and sends nothing anywhere.)
Install
Copy the installer prompt below into your Muse — your agent does the rest.

Version:

Quick answer

Taste Skill — Mobile Image Generation is a free Muse skill, an art-direction skill that turns Muse into a mobile product design director. Describe an app screen or flow, such as onboarding, dashboard, chat or checkout, and it generates premium, app-native screen concepts with coherent style, controlled palettes and believable phone framing.

⌁

Install

Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Oct 19, 2026📌 Pinned version: commit beb16b0

MOVED REPOSITORY — READ THIS FIRST: this skill has moved. It used to live in leonxlnx/taste-skill (skills/imagegen-frontend-mobile/SKILL.md), where what remains at that path is now only a 468-character stub: "Moved to github.com/Leonxlnx/imagegen-skills". The stub carries disable-model-invocation: true (it runs only when called by name) and directs you to install from the new repository with: npx skills add https://github.com/Leonxlnx/imagegen-skills --skill "imagegen-frontend-mobile". The install name did not change: it is still "imagegen-frontend-mobile". This prompt follows the stub's direction by direct pinned fetch from the new repository, below — do not install the old taste-skill stub as the skill. Step 0, before installing anything: read this install prompt, summarize in a few lines what this build does and what installing it involves — including the repository move above — then ask me to confirm before you proceed. Prerequisites: none for installation — no account, API key, or paid plan required, and this skill is free. (Using the skill later requires a host agent with image-generation capability, for example ChatGPT Images, Codex image mode, or any agent that can generate images; installing the skill itself requires none of that and sends nothing anywhere.) Install "imagegen-frontend-mobile" for me — an elite mobile app image-generation skill by @leonxlnx, MIT-licensed, in its new home Leonxlnx/imagegen-skills. It generates images only: premium, app-native mobile screen concepts and screen-flow images for iOS, Android, and cross-platform products — onboarding flows, auth flows, home dashboards, profile screens, settings, chat, ecommerce, fintech, health and fitness, productivity, social apps, utilities, multi-screen app concepts, and premium mobile redesigns. It does not write code, and it is not for websites, landing pages, desktop dashboards, image-to-code, or frontend implementation: it must not switch into coding mode, describe code, or build SwiftUI, React Native, Flutter, or HTML. How it works, from the file: decide the platform mode first (iOS-native premium, Android-native premium, or cross-platform premium neutral — never mixing iOS and Android patterns carelessly), then generate the screen image or screen set directly, generating enough screens to make the flow real (a requested flow gets separate screens, never one compressed collage board with tiny text; a detail view is generated fresh, never cropped out of an older larger image). For a multi-screen set it locks an internal design bible first (platform mode, device frame, palette, typography, spacing, corner radii, icon style, imagery, texture, navigation, components) and keeps every screen in that one product world while varying composition and tempo, in a logical screen-to-screen progression. By default each screen is shown inside a subtle, clean, premium phone mockup with a visible frame, evenly padded and content-first — the frame supports the screen, never dominates it, and is omitted only when I explicitly ask for raw screen-only output. Its baseline dials (design variance 8, visual density 3, art direction 9, platform awareness 9, flow variety 8, spacing generosity 9, texture strength 7, and 10 for clarity discipline, image-generation eagerness, color-palette discipline, non-genericity, consistency strength, flow-logic discipline, text-readability priority, content-first mockup balance, and minimum text-size discipline) drive a style-variation engine (one theme paradigm, typography character, structure bias, image art-direction bias, texture treatment, and palette logic, plus exactly four signature components, two decorative asset sets, and two motion-implied cues), category-specific biases (for example fintech: trust, calm spacing, clear numbers, no fake chart spam; social: profile and feed rhythm, stronger flow variety), strict anti-AI-tell rules (no purple-blue gradient clichés, random glass cards, ambient blobs, cloned screens, filler copy like "elevate your life" or fake brands like Acme or NovaCore, no generic developer-tool / Lucide-like icon defaults, no pills-and-badges clutter, no tiny unreadable text), and a regeneration rule: a weak screen — too small text, fake navigation, website-like layout, muddy palette, sloppy mockup — is regenerated, not settled for, against a 27-point internal quality check before output. Repository: https://raw.githubusercontent.com/Leonxlnx/imagegen-skills/beb16b0d0f4dae9a6f5cb04ea41e97b93a54ab8e/skills/imagegen-frontend-mobile/SKILL.md (the skill is the single SKILL.md file in the skills/imagegen-frontend-mobile/ folder of the Leonxlnx/imagegen-skills repository at that commit) Pinned version: install the code at commit beb16b0d0f4dae9a6f5cb04ea41e97b93a54ab8e exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. [FINGERPRINTS] SKILL.md at the pinned commit: SHA-256 8a33389979f3074fa0926678e266ad2eb9234624472254469fc1ad916b9caa24, 40,326 bytes. If the fetched file does not match, it changed since the review: stop, tell me what you found and let me decide. Updated October 19, 2026: Skill Harbor regenerated this install prompt from the current upstream file because the skill moved repositories. Previous state: this listing pointed to leonxlnx/taste-skill at commit 717446e07a78d4e6d7918b1fde1a263378eebac9, where skills/imagegen-frontend-mobile/SKILL.md was the full skill. That repository path is now only the 468-character move stub described above (disable-model-invocation: true, so it runs only when called by name, pointing to Leonxlnx/imagegen-skills and giving the npx skills add command with the unchanged install name "imagegen-frontend-mobile"). The full skill now lives in Leonxlnx/imagegen-skills at the new pinned commit above, as the single SKILL.md file in skills/imagegen-frontend-mobile/. Corrections to the previous prompt: its repository and pin are superseded (install from the new repository, not from taste-skill); and its generic safety boilerplate — that the repo "should contain zero secrets in code, credentials only via the secure vault, allowed hosts declared in the SKILL.md" — does not describe this file: the new SKILL.md declares no allowed hosts, contains no code and no scripts, and asks for no credentials at all, as confirmed in the safety check below. What the skill does is otherwise the same art direction the previous prompt described: premium, readable, consistent mobile screen images, in a phone mockup by default, images only and no code. 1. Fetch SKILL.md from the repository URL above at the pinned commit into a temporary folder, then compare it with the fingerprint from Skill Harbor's review of 2026-10-19 (see [FINGERPRINTS] above): - SKILL.md: SHA-256 8a33389979f3074fa0926678e266ad2eb9234624472254469fc1ad916b9caa24 If it does not match, it changed since the review: do not install it yet, tell me what you found and let me decide. 2. Safety check: read SKILL.md and review it for anything suspicious (unexpected network calls, shell commands, credential harvesting, code that runs at install); STOP on any red flag and tell me. Tell me in plain words what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Confirm these points from the file: - Nothing is sent anywhere during installation itself, and nothing runs during installation: the skill is a single Markdown instruction file (SKILL.md) — its folder holds no scripts, no helper files, and no executable code at all. - At use, the file itself sends nothing and names no external service, API, or host: it instructs the host agent to generate mobile screen images and flows with the host's own image-generation capability, and the images are its only output. It generates images only — it does not write, describe, or build code of any kind. Because generation is done by the host, do not put personal information in the app descriptions you hand it for generation. - No part of the file reads credentials, API keys, or personal information, and nothing in it asks for any: there are no network instructions, no shell commands, and no data collection in the file. - The working rules the file imposes, as summarized above: platform mode chosen first and held coherently; the screen set generated directly and in sufficient number, with fresh detail renders instead of crops; a locked design bible and multi-screen consistency; a logical flow; a clean, visible, content-first phone mockup by default; comfortably readable text that is never shrunk to fit (split or regenerate the screen instead); respect for safe areas and system regions; texture, imagery behind text (with fades or masks protecting readability), creative assets, and custom-feeling iconography used deliberately; and weak screens regenerated against its 27-point quality check. 3. When I say yes, install it as a skill: copy SKILL.md into the agent's skills directory, in a folder named "imagegen-frontend-mobile". Do not install the old taste-skill move stub as a separate skill. 4. Verify with no network calls: frontmatter valid (name: imagegen-frontend-mobile, description present), the file in place, and its SHA-256 matching the fingerprint above. 5. Report what was installed, where, and what I still need to do myself (e.g. start a fresh session so the host re-indexes the skill; then simply describe the app screens or flow I want generated, using a host agent with image-generation capability). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.

🔒 Security, provenance and what this touches
🧭 What this touches
Based on the install prompt we scanned, this skill does not run commands, make third-party network calls, install packages, write files, or ask for API keys.
Install prompt analysis · 2026-10-13
🔒 Security and provenance
📌 Pinned version: commit beb16b0 · 2026-10-19
🧪 Install ran in our test sandbox · Oct 19, 2026
👁 Observed during the install test · Oct 19, 2026 : no network contact
🔎 Last checked · Oct 19, 2026 : no change detected in the files linked by this install prompt
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Wanna be extra careful?

The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?

How to check a skill before installing →

✓

Verification certificate

Dated facts about what was checked for this listing, and how. This is not a safety certification.

Pinned version
beb16b0 · 2026-10-19
Sandbox install test
Install ran in our test sandbox · Oct 19, 2026
Creator identity
Not verified
Fingerprint watch
Last checked · Oct 19, 2026 : no change detected in the files linked by this install prompt
Install prompt analysis
2026-10-13 : No commands, network calls, packages, file writes or API keys detected.

Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify

?

Questions

How do I install a skill?

Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.

Where does my money go?

Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.

What does the ✓ next to a creator’s name mean?

It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.