Photoshop Cutout logo

Photoshop Cutout

Organization
THU-SAGE
photoshop-cutout

Remove the background from an image and produce a transparent PNG by driving the real Adobe Photoshop app on a macOS host. Triggers: remove background, cut out, knock out the subject, isolate the subject, make a transparent PNG, transparent background, delete background. 中文触发词:抠图、扣图、去背景、去掉背景、扣出主体、透明背景、做透明 PNG。

Overview

PublisherTHU-SAGE
Repositorysyll
Skill namephotoshop-cutout
Stars
303
Forks
2
Bundled files
Instructions only
LicenseMIT
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.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by THU-SAGE on GitHub. Read the source before you install it.

Installation

Install the Photoshop Cutout 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/THU-SAGE/syll/tree/main/syll/skills/photoshop-cutout
  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/THU-SAGE/syll.git /tmp/syll
mkdir -p .claude/skills
cp -r /tmp/syll/syll/skills/photoshop-cutout .claude/skills/photoshop-cutout
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Photoshop Cutout 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 Photoshop Cutout 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 Photoshop Cutout 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.

Photoshop Cutout

Use the photoshop_cutout tool to remove the background from an image and deliver a transparent PNG. This is not a filter that runs in this process — it drives the real Adobe Photoshop application on a macOS host, opens the image, runs the cutout, and exports the result.

When To Reach For This

The user wants a subject lifted off its background. Recognize the intent from phrases like:

  • English: "remove the background", "cut out the person/product/logo", "knock out the subject", "isolate this", "make it a transparent PNG", "give me a PNG with no background".
  • 中文:「抠图」「扣图」「帮我去背景」「去掉背景」「把主体扣出来」「要透明背景」「做成透明 PNG」。

How It Works

  1. You call photoshop_cutout with the path to the source image.
  2. The tool launches / focuses Adobe Photoshop, opens the image, performs the subject-selection cutout, and exports a transparent PNG.
  3. The tool measures the result (how much background was actually removed, whether a clean alpha channel was produced) and returns a verdict.
  4. The before image and the after PNG render inline automatically — you do not need to attach or describe them yourself.

Confirm Before Control

This tool seizes the mouse and keyboard of the host machine. It MUST NOT take over the screen without the user's explicit permission.

  1. First call — confirmed=false. Always make the first call with confirmed=false. The tool will return a takeover-consent question (it does not touch the mouse/keyboard yet). Surface that question to the user.
  2. Wait for an explicit yes. Only proceed once the user clearly agrees — a "yes", "go ahead", "do it", "确认", "可以" in the conversation counts. Silence, ambiguity, or "maybe" does not count.
  3. Second call — confirmed=true. Only then call again with confirmed=true. This is the call that actually takes over the host.

Never set confirmed=true on the first call, and never assume consent.

Honesty Protocol

Report only what the tool's measured verdict supports. Do not embellish.

The tool returns a Markdown report whose first line is either **Verdict**: PASS … or **Verdict**: REVIEW …, followed by measured numbers (transparent %, subject %, whether an alpha channel is present).

  • PASS — the alpha checks passed: you may say the background was removed and a transparent PNG was produced.
  • REVIEW — the file was exported but not all transparency checks passed. Say the cutout needs review / is not fully clean — do not claim the background was removed. Offer to retry or refine.
  • Export failed (the verdict text says the cutout PNG was not created, shown as REVIEW) — say it failed and offer to retry or try a different image.

Never claim "done", "cut out", or "transparent" beyond what the verdict and the measured ratios support. The user trusts the numbers, not your optimism.

Usage Protocol

When the user asks for a cutout:

  1. Confirm the source. Make sure you have the path to the image the user wants processed.
  2. First call, confirmed=false. Relay the takeover-consent question.
  3. Get explicit consent. Wait for the user's "go ahead".
  4. Second call, confirmed=true. Run the real cutout.
  5. Report the verdict honestly. State the outcome exactly as measured. The before/after images render inline on their own.

Example

User: "帮我把这张产品图抠图,要透明背景"

  • First, call photoshop_cutout(image_path="/path/to/product.jpg", confirmed=false).
  • The tool replies with a consent question; relay it: "This will take over the mouse and keyboard on the Mac to run Photoshop. Shall I go ahead?"
  • User: "可以,去吧"
  • Then call photoshop_cutout(image_path="/path/to/product.jpg", confirmed=true).
  • Tool returns **Verdict**: PASS. You report: "背景已去除,导出了透明 PNG(结果已通过校验)。" The before/after images appear inline. If it returns REVIEW instead, say the cutout needs another look rather than claiming success.

Do Not

  • Do not drive gui_action directly to attempt the cutout — use this dedicated tool, which knows the Photoshop workflow and measures the result.
  • Do not call with confirmed=true before the user agrees.
  • Do not overstate the result. A REVIEW verdict is not a clean cutout.

Frequently asked questions

What does the Photoshop Cutout AI skill do?

Remove the background from an image and produce a transparent PNG by driving the real Adobe Photoshop app on a macOS host. Triggers: remove background, cut out, knock out the subject, isolate the subject, make a transparent PNG, transparent background, delete background. 中文触发词:抠图、扣图、去背景、去掉背景、扣出主体、透明背景、做透明 PNG。

Why use Photoshop Cutout on TypingMind?

Because you install it once and use it with any model. Photoshop Cutout 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 Photoshop Cutout in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/THU-SAGE/syll/tree/main/syll/skills/photoshop-cutout. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Photoshop Cutout?

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 Photoshop Cutout?

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

Is the Photoshop Cutout AI skill free?

Yes. It is published on GitHub by THU-SAGE under the MIT license. 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 👇