App Icon Template Creator logo

App Icon Template Creator

Community
xzhih
app-icon-template-creator

Use when one or more Apple app icon references, generated concepts, screenshots, SVGs, or icon packages need to become a reusable app-icon-gen style template.

Overview

Publisherxzhih
Repositoryapp-icon-skill
Skill nameapp-icon-template-creator
Stars
14
Forks
2
Bundled files
10
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 10 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by xzhih on GitHub. Read the source before you install it.

Installation

Install the App Icon Template Creator AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/xzhih/app-icon-skill/tree/main/skills/app-icon-template-creator
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/xzhih/app-icon-skill.git /tmp/app-icon-skill
mkdir -p .claude/skills
cp -r /tmp/app-icon-skill/skills/app-icon-template-creator .claude/skills/app-icon-template-creator
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable App Icon Template Creator in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads App Icon Template Creator on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so App Icon Template Creator is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

App Icon Template Creator

Create a reusable app-icon-gen style template from reference icon artwork. The goal is not to copy a source icon. The goal is to extract a transferable design language that can generate different app icons with a consistent style and then package cleanly through app-icon-gen.

REQUIRED SUB-SKILLS: Use app-icon-gen and an available raster image-generation skill such as imagegen or codex-image-gen. If either capability is unavailable before the workflow can run, report dependency_blocked; do not call the template complete. Use needs_package_verification only when those capabilities are available but the selected migration candidate has not yet passed the package flow.

Hard Contract

  • Inspect supplied references before asking questions: image files, SVGs, existing .icon bundles, generated concepts, screenshots, or project assets.
  • Use one working root for each template creation run: <delivery>/tmp/work/.
  • Preserve the selected source reference as <delivery>/<template-name>.origin.png or <delivery>/<template-name>.origin.svg.
  • Record the current analysis in <delivery>/tmp/work/reference-analysis.md.
  • Record each reconstruction prompt in <delivery>/tmp/work/reconstruction-prompt-<n>.txt.
  • Save reconstruction outputs under <delivery>/tmp/work/reconstruction-<n>.png.
  • Record each migration-test prompt in <delivery>/tmp/work/migration-<n>-prompt.txt.
  • Save migration-test outputs under <delivery>/tmp/work/migration-<n>.png.
  • Record reconstruction and migration decisions in <delivery>/tmp/work/reconstruction-review.md and <delivery>/tmp/work/migration-review.md.
  • Write the final candidate template to <delivery>/<template-name>.md.
  • Keep the final candidate template compatible with app-icon-gen/examples/semantic-image-prompts/: do not embed reconstruction or migration evidence sections in the final template.
  • Do not call the template complete until reconstruction, migration, and package-readiness gates have all been checked.
  • Do not create a template that preserves a proprietary logo, exact mascot identity, unique brand mark, or exact source arrangement unless the user explicitly owns it and asks for a private template.

Workflow

  1. Choose a delivery folder and template slug. Use a short lowercase hyphenated slug such as black-stage-pane-cluster.
  2. Read references/reference-analysis-protocol.md and analyze the source icon through that lens.
  3. Read references/reconstruction-gates.md, then write a reconstruction prompt that should reproduce the source style without naming a reusable template yet.
  4. Use the available raster image-generation skill to generate one or more reconstruction images from the prompt. When a source image is supplied, pass it as a style/reference image and label its role in the prompt.
  5. Compare source and reconstruction with visual inspection and, when useful, app-icon-gen reference comparison reports. Repair the analysis and prompt until the reconstruction is style-equivalent or the limitation is documented in tmp/work/reconstruction-review.md.
  6. Before extracting the template, show the reconstruction candidates to the user when the user is actively reviewing the run. Continue with the selected/accepted reconstruction; if no live review is available, record the agent's acceptance basis in tmp/work/reconstruction-review.md.
  7. Read references/template-extraction.md and extract the reusable style grammar from the successful reconstruction: fixed style constraints, variant axes, app-specific subject slots, conversion flow, and drift guards.
  8. Read references/migration-tests.md, then generate at least three migration tests for different app contexts. Use contexts that change the app metaphor while keeping the style family.
  9. Show migration outputs to the user when the user is actively reviewing the run. Continue with selected/accepted candidates; if no live review is available, record the agent's acceptance basis in tmp/work/migration-review.md.
  10. Judge whether the migration outputs still feel like the same template. If they diverge, repair the template constraints before continuing.
  11. Read references/template-output-format.md, then write the final candidate template Markdown file.
  12. Run npm --prefix <skill-dir>/scripts ci, then npm --prefix <skill-dir>/scripts run template-lint -- <candidate-template.md>, resolving <skill-dir> to this skill's installed directory.
  13. For at least one user-selected or accepted migration output, run the app-icon-gen package flow and verification. Use the candidate template's conversion flow as the route. If the package cannot become a complete candidate, revise the template.

Reference Routing

  • references/reference-analysis-protocol.md: always read first; defines the image-analysis lens.
  • references/reconstruction-gates.md: read before using the available raster image-generation skill to reproduce the source style.
  • references/template-extraction.md: read after a reconstruction is accepted.
  • references/migration-tests.md: read before testing the draft template across multiple app concepts.
  • references/template-output-format.md: read before writing or linting the final template.

Completion Gate

Complete only when all applicable checks are true:

  • A source evidence file exists at the delivery root.
  • tmp/work/reference-analysis.md exists and names the style invariants and drift risks.
  • At least one reconstruction output was generated and inspected.
  • tmp/work/reconstruction-review.md documents the accepted reconstruction, reviewer basis, and any non-template-critical differences.
  • At least three migration outputs were generated and inspected.
  • tmp/work/migration-review.md documents at least three migration concepts and whether each preserved the same design language while changing app-specific subject semantics.
  • The final template contains all required sections from template-output-format.md.
  • template-lint passes.
  • At least one generated migration candidate has been packaged through app-icon-gen and reached complete_candidate. If packaging cannot run, return needs_package_verification instead of completing the template.

Final response must include the delivery path, final template path, source evidence path, reconstruction outputs inspected, migration outputs inspected, template-lint result, package verification result, and any accepted limitations.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the App Icon Template Creator AI skill do?

Use when one or more Apple app icon references, generated concepts, screenshots, SVGs, or icon packages need to become a reusable app-icon-gen style template.

Why use App Icon Template Creator on TypingMind?

Because you install it once and use it with any model. App Icon Template Creator is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install App Icon Template Creator in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/xzhih/app-icon-skill/tree/main/skills/app-icon-template-creator. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use App Icon Template Creator?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with App Icon Template Creator?

As many as you like. As long as a model supports skills, you can use App Icon Template Creator with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the App Icon Template Creator AI skill free?

It is published on GitHub by xzhih. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇