agentCLAWHUBUnverified

DesignKit Ecommerce AI Studio

Use when users need DesignKit image editing, ecommerce product-image generation, or A+ detail-page planning and rendering, with public metadata optimized for...

OpenClaw

Rank

62

Safety

84

Downloads

2.0k

Updated

Oct 9, 2026

Version

1.0.6

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 2K downloads reported by the source. Last updated 10/9/2026.

Avoid when

  • Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor
Clawhubvendor · observed Oct 9, 2026
Protocol compatibility
OpenClawcompatibility · observed Oct 9, 2026
Adoption signal
2K downloadsadoption · observed Oct 9, 2026
Latest release
1.0.6release · observed May 6, 2026
Handshake status
UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s1774n8e9dc5jhjx29yevfhptd83emt0:designkit-skills
  1. Install using `clawhub skill install s1774n8e9dc5jhjx29yevfhptd83emt0:designkit-skills` in an isolated environment before connecting it to live workloads.
  2. No published capability contract is available yet, so validate auth and request/response behavior manually.
  3. Review the upstream CLAWHUB listing at https://clawhub.ai/meituskills/designkit-skills before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-meituskills-designkit-skills/snapshot"

Documentation

CLAWHUB

150,733 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: designkit-skills
description: Use when users need DesignKit image editing, ecommerce product-image generation, or A+ detail-page planning and rendering, with public metadata optimized for ClawHub and runtime guidance defaulting to Simplified Chinese.
version: "1.0.0"
metadata:
  openclaw:
    requires:
      env:
        - DESIGNKIT_OPENCLAW_AK
      bins:
        - bash
        - curl
        - python3
    primaryEnv: DESIGNKIT_OPENCLAW_AK
    homepage: https://www.designkit.cn/openclaw
---

# DesignKit Skills

Public ClawHub bundle for DesignKit image editing and ecommerce product-image workflows.
实际运行时默认面向中文用户,用简体中文收集参数、确认操作并返回结果。

## Purpose

- Route to `designkit-edit-tools` for single-image background removal or image restoration.
- Route to `designkit-ecommerce-product-kit` for multi-step ecommerce product-image generation.
- Route to `designkit-ecommerce-a-plus-detail` for A+ detail-page planning and detail-image generation.
- Keep the public package description English-friendly for ClawHub, while keeping the actual user conversation natural in Simplified Chinese.

## Public Installation Posture

- Explain actions in product terms such as background removal, image restoration, or ecommerce image generation.
- Default user-facing replies to Simplified Chinese unless the user explicitly prefers another language.
- Only process image URLs or local file paths that the user explicitly provides for the requested task.
- If the user supplies a local image path, make clear that the file will be uploaded to the remote DesignKit / OpenClaw service.
- Do not expose credentials, raw request or response payloads, internal headers, or local script paths unless the user explicitly asks for technical details.

## Routing Rules

### 1. `designkit-edit-tools`

Route here when user intent matches:

- Background removal, cutout, transparent background, white background
- Image restoration, blurry photo fix, image enhancement, sharpening

Common search terms:
`抠图`, `去背景`, `透明底`, `白底图`, `画质修复`, `变清晰`, `image restoration`, `background removal`

### 2. `designkit-ecommerce-product-kit`

Route here when user intent matches:

- Ecommerce listing image sets
- Product hero images and detail images
- Launch-ready marketplace image packs from an existing product photo

Common search terms:
`电商套图`, `商品套图`, `主图详情图`, `listing images`, `amazon listing images`, `product image set`

### 3. `designkit-ecommerce-a-plus-detail`

Route here when user intent matches:

- A+ detail page planning
- Product detail-page concept and module strategy
- A+ detail-image generation from 1-3 product images

Common search terms:
`A+详情`, `详情页`, `A+页面`, `detail page`, `A+ content`, `详情图`, `Amazon A+`

If the request is ambiguous, ask one short clarification question before routing.

## Conversation Flow (Required)

1. Recognize the user intent and route to the matching sub-skill.
2. Ask for the missing image first if the user has not already provided one.
3. Ask only the 1-2 most imp

skills/designkit-ecommerce-a-plus-detail/SKILL.md

---
name: designkit-ecommerce-a-plus-detail
description: Use when users want A+ detail-page planning and detail-image generation from 1-3 product images, with runtime guidance defaulting to Simplified Chinese.
version: "1.0.0"
metadata:
  openclaw:
    requires:
      env:
        - DESIGNKIT_OPENCLAW_AK
      bins:
        - bash
        - curl
        - python3
    primaryEnv: DESIGNKIT_OPENCLAW_AK
    homepage: https://www.designkit.cn/openclaw
---

# DesignKit Ecommerce A+ Detail

A+ 详情页双阶段工作流:先生成详情策划方案,再生成详情图。
公开元数据可偏英文或双语,运行时分步提问与结果说明默认使用简体中文。

## User Quick Start (CN)

给终端里的 Agent 直接说下面任意一句即可:

1. 最短说法

`帮我做 A+ 详情页,这是产品图:https://example.com/p1.jpg`

2. 常用完整说法

`帮我做 A+ 详情页。产品图:https://example.com/p1.jpg, https://example.com/p2.jpg;商品卖点与要求:轻量、防水、适合通勤;平台/市场/语言/比例:亚马逊/美国/英文/970:600;模块:首屏主视觉、核心卖点图、使用场景图。`

3. 使用本地图片路径

`帮我做 A+ 详情页,产品图在 /path/to/product.jpg`

说明:使用本地路径时,图片会上传到 DesignKit / OpenClaw 远程服务处理。

## User Input Contract

- 产品图:1-3 张(URL 或本地路径)
- 必填:商品卖点与要求(`product_info`)
- 可选:平台、市场、语言、比例、模块

如果用户不填可选项,默认值:

- `platform`: `amazon`
- `market`: `US`
- `language`: `English`
- `aspect_ratio`: `970:600`
- `selected_modules`: `首屏主视觉,核心卖点图,使用场景图,多角度图,场景氛围图,商品细节图`

## User Interaction Expectation

Agent 与用户对话应保持 3 步即可完成:

1. 先收产品图(1-3 张)
2. 再收商品卖点与要求
3. 最后确认配置(平台/市场/语言/比例/模块),未提供则走默认值

## Public Installation Posture

- Explain this capability in product terms such as A+ detail page, detail-page plan, or detail image set.
- Only process image URLs or local file paths explicitly provided by the user.
- If local image paths are provided, clearly indicate that those images will be uploaded to DesignKit / OpenClaw.
- Do not expose credentials, raw payloads, internal headers, or local script paths unless explicitly requested.

## Required Flow

### 1. Product Images (1-3)

- Ask for product images first when missing.
- Accept either URL(s) or local image path(s).
- The image count must be between 1 and 3.
- If more than 3 images are provided, ask the user to reduce to 3 or fewer before execution.

### 2. Product Selling Points And Requirements

After images are received, ask for `product_info` (商品卖点与要求) before execution.

### 3. Configuration

Then collect or confirm these fields:

- `platform`
- `market`
- `language`
- `aspect_ratio`
- `selected_modules`

Defaults if user skips:

- `platform`: `amazon`
- `market`: `US`
- `language`: `English`
- `aspect_ratio`: `970:600`
- `selected_modules`: `首屏主视觉,核心卖点图,使用场景图,多角度图,场景氛围图,商品细节图`

### 4. Two-Step Execution (Required)

1. Submit and poll detail-page planning.
2. Show the planned modules to the user.
3. Generate detail images only after plan output is available (and user confirmation if needed).

Note:

- `detail_plan_submit` uses `step=3` internally (fixed).

### 5. Regen

If the user requests regeneration of a specific generated image, call regen with `task_id`, then continue poll.

## Commands

```bash
bash __SKILL_DIR__/../../scripts/run_ecommerce_a_plus_detail.sh detail_plan_submit --input-json '<j

skills/designkit-ecommerce-product-kit/SKILL.md

---
name: designkit-ecommerce-product-kit
description: Use when users want ecommerce listing images, product hero/detail images, or a multi-step product-image workflow from an existing product photo, with runtime guidance defaulting to Simplified Chinese.
version: "1.8.0"
metadata:
  openclaw:
    requires:
      env:
        - DESIGNKIT_OPENCLAW_AK
      bins:
        - bash
        - curl
        - python3
    primaryEnv: DESIGNKIT_OPENCLAW_AK
    homepage: https://www.designkit.cn/openclaw
---

# DesignKit Ecommerce Product Kit

Multi-step ecommerce image workflow for product hero images and detail images.
公开元数据可偏英文或双语,但分步提问、确认和结果说明默认使用简体中文。

## Public Installation Posture

- Explain the workflow in product terms such as listing image set, hero image, detail image, or marketplace pack.
- Only process product image URLs or local file paths that the user explicitly provides.
- If a local product image path is provided, make clear that the file will be uploaded to the remote DesignKit / OpenClaw service.
- Do not expose credentials, raw request or response payloads, internal headers, or local script paths unless the user explicitly asks for technical details.

## Required Flow

### 1. Product Image

If the user has not provided a product image, ask for one URL or local image path first.

### 2. Selling Points And Style Preference

After receiving the product image:

- ask only for selling points and style preference
- do not ask for platform, country, language, or ratio in the same reply
- if useful, offer a suggested selling-point summary the user can accept or edit
- ask this step in Simplified Chinese

### 3. Listing Configuration

Only after the user responds to step 2, ask for:

- platform
- market / country
- language
- aspect ratio

Default values if the user skips this step:

- platform: `amazon`
- market: `US`
- language: `English`
- aspect ratio: `1:1`

If the user already provided config together with the selling-point reply, accept it and skip this extra question.
Ask this configuration step in Simplified Chinese as well.

### 4. Style Selection Is Optional

Only enter the style-selection flow when the user clearly asks for a specific style direction or wants to review style options.

- if the user says `没有`, `随便`, `自动`, or gives no style direction, skip style selection
- do not invent styles yourself
- if style options are shown, the final chosen `brand_style` must come directly from the API response

### 5. Render And Download

After the staged inputs are complete:

1. submit the render task
2. poll until complete
3. download the generated images locally
4. show the result images and tell the user where they were saved

If the user points to a specific generated image and asks for regeneration, call `render_regen` with `transfer_id` and `task_id`, then continue with `render_poll`.

## API Value Mapping

| User-facing choice | API value |
| --- | --- |
| 亚马逊 | `amazon` |
| 淘宝 | `taobao` |
| Temu | `temu` |
| 拼多多 | `pinduoduo` |
| Tik

skills/designkit-edit-tools/SKILL.md

---
name: designkit-edit-tools
description: Use when users want background removal, transparent or white-background output, or image restoration from a single image, while keeping runtime replies in Simplified Chinese.
version: "1.0.0"
metadata:
  openclaw:
    requires:
      env:
        - DESIGNKIT_OPENCLAW_AK
      bins:
        - bash
        - curl
        - python3
    primaryEnv: DESIGNKIT_OPENCLAW_AK
    homepage: https://www.designkit.cn/openclaw
---

# DesignKit Edit Tools

Single-image editing skill for background removal and image restoration.
公开说明可偏英文或双语,但实际用户对话默认使用简体中文。

## Public Installation Posture

- Explain the action in plain product terms such as cutout, transparent background, white background, or image enhancement.
- Only use image URLs or local file paths that the user explicitly provided.
- If the user gives a local path, make clear that the file will be uploaded to the remote DesignKit / OpenClaw service.
- Do not expose credentials, raw JSON payloads, internal headers, or local script paths unless the user explicitly asks for technical details.

## Intent Mapping

| User says | Action |
| --- | --- |
| 抠图、去背景、背景移除、透明底、白底图、matting、background removal | `sod` |
| 变清晰、画质修复、图片增强、超分、image restoration | `image_restoration` |

## Conversation Flow

- If no image is provided, ask for one image path or URL first.
- Do not ask for optional settings unless the user explicitly cares about them.
- Keep the confirmation short, then execute immediately in the same turn.
- User-facing confirmations and error guidance should default to Simplified Chinese.

Example confirmations:

- `好的,我来帮你把这张图抠成透明底。`
- `好的,我来帮你提升这张图的清晰度。`

## Execution

```bash
bash __SKILL_DIR__/../../scripts/run_command.sh <action> --input-json '<params_json>'
```

Examples:

```bash
bash __SKILL_DIR__/../../scripts/run_command.sh sod --input-json '{"image":"https://example.com/photo.jpg"}'
```

```bash
bash __SKILL_DIR__/../../scripts/run_command.sh image_restoration --input-json '{"image":"/Users/me/photo.jpg"}'
```

These commands are internal execution guidance for the agent. Do not quote them to end users unless they explicitly ask for implementation details.

## Runtime And Safety

- Requires `DESIGNKIT_OPENCLAW_AK`.
- Local uploads are limited to `JPG/JPEG/PNG/WEBP/GIF` image files.
- Local images may be uploaded to the remote DesignKit / OpenClaw API.
- Request logging is off by default. If `OPENCLAW_REQUEST_LOG=1` is enabled for debugging, credentials and signed upload fields stay redacted.

## Privacy Defaults

- Request logging is disabled by default.
- Sensitive fields remain redacted if logging is enabled manually.
- Local uploads happen only when the user explicitly provides a local image path.
- Default runtime language is Simplified Chinese.

## Result Handling

- `ok: true`: show the returned `media_urls`.
- `ok: false`: show `user_hint` and do not dump raw JSON.

## Error Guide

| `error_type` | User-facing action |
| --- | --- |
| `CREDENTIALS_MISS

README.md

# designkit-skills

Public DesignKit skill bundle for OpenClaw / ClawHub.
面向中文用户的 DesignKit 技能包,支持抠图、画质修复、电商商品套图和 A+ 详情页。

It covers four common image-production tasks:

- background removal for transparent or white-background outputs
- image restoration for blurry or low-quality photos
- ecommerce listing-image generation from an existing product photo
- A+ detail-page planning and detail-image generation from 1-3 product images

## Public Package Positioning / 对外发布定位

- Public metadata and marketplace copy are written in English or bilingual form so ClawHub can index, review, and display the package more reliably.
- Actual user-facing prompts, confirmations, and result guidance should default to Simplified Chinese unless the user explicitly prefers another language.
- This package is intended for Chinese-language ecommerce and image-editing workflows, but keeps a ClawHub-friendly public surface.

## What It Does / 功能概览

- removing backgrounds from product photos
- generating transparent PNG / clean cutout-style assets
- sharpening blurry images before reuse
- creating hero images and detail images for ecommerce listings
- planning A+ detail modules and rendering detail-page image sets

## Included Skills

- `designkit-skills`: root router for the public bundle
- `designkit-edit-tools`: single-image background removal and restoration
- `designkit-ecommerce-product-kit`: multi-step ecommerce listing-image workflow
- `designkit-ecommerce-a-plus-detail`: two-step A+ detail workflow (planning -> rendering)

## Install

Inspect first:

```bash
clawhub inspect designkit-skills
```

Install from ClawHub:

```bash
clawhub install designkit-skills
```

Install directly from GitHub:

```bash
npx -y skills add https://github.com/meitu/designkit-skills
```

## Required Environment

Get an API key from [DesignKit OpenClaw](https://www.designkit.cn/openclaw), then set:

```bash
export DESIGNKIT_OPENCLAW_AK="AK"
```

## Runtime Language / 交互语言

- Default user-facing language: Simplified Chinese
- Switch to English only when the user explicitly asks or clearly uses English throughout the conversation
- Keep public metadata English-friendly, but keep runtime guidance natural for Chinese users

## Runtime Behavior / 运行行为

- local image paths may be uploaded to the remote DesignKit / OpenClaw API for processing
- ecommerce renders are downloaded back to a local output directory after completion
- A+ detail renders are downloaded back to a local output directory after completion
- request logging is disabled by default
- if `OPENCLAW_REQUEST_LOG=1` is enabled for debugging, credentials and signed upload data stay redacted
- local uploads only accept `JPG/JPEG/PNG/WEBP/GIF` image files

## Public Safety Posture / 安全说明

- Only image URLs or local file paths explicitly provided by the user should be processed.
- Do not browse unrelated local directories or infer local file paths on behalf of the user.
- If a local image path is used, the skill should clearly indicate 
Github ReposUpdated 10h agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW
Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW
Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.

MCPOPENCLAW
Github ReposUpdated 7mo agoRank 70

CopilotKit

The Frontend for Agents & Generative UI. React + Angular

OPENCLAW

Machine-readable data

The same record, as JSON, for agents and crawlers.

{
  "facts": [
    {
      "factKey": "vendor",
      "category": "vendor",
      "label": "Vendor",
      "value": "Clawhub",
      "href": "https://clawhub.ai/meituskills/skills/designkit-skills",
      "sourceUrl": "https://clawhub.ai/meituskills/skills/designkit-skills",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-09T20:43:21.097Z",
      "isPublic": true
    },
    {
      "factKey": "protocols",
      "category": "compatibility",
      "label": "Protocol compatibility",
      "value": "OpenClaw",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-meituskills-designkit-skills/contract",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-meituskills-designkit-skills/contract",
      "sourceType": "contract",
      "confidence": "medium",
      "observedAt": "2026-10-09T20:43:21.097Z",
      "isPublic": true
    },
    {
      "factKey": "traction",
      "category": "adoption",
      "label": "Adoption signal",
      "value": "2K downloads",
      "href": "https://clawhub.ai/meituskills/designkit-skills",
      "sourceUrl": "https://clawhub.ai/meituskills/designkit-skills",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-09T20:43:21.097Z",
      "isPublic": true
    },
    {
      "factKey": "latest_release",
      "category": "release",
      "label": "Latest release",
      "value": "1.0.6",
      "href": "https://clawhub.ai/meituskills/designkit-skills",
      "sourceUrl": "https://clawhub.ai/meituskills/designkit-skills",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-05-06T03:21:38.944Z",
      "isPublic": true
    },
    {
      "factKey": "handshake_status",
      "category": "security",
      "label": "Handshake status",
      "value": "UNKNOWN",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-meituskills-designkit-skills/trust",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-meituskills-designkit-skills/trust",
      "sourceType": "trust",
      "confidence": "medium",
      "observedAt": null,
      "isPublic": true
    }
  ],
  "events": [
    {
      "eventType": "release",
      "title": "Release 1.0.6",
      "description": "* 新增 `designkit-ecommerce-a-plus-detail` 子技能,支持 A+ 商品详情页规划与图片生成。",
      "href": "https://clawhub.ai/meituskills/designkit-skills",
      "sourceUrl": "https://clawhub.ai/meituskills/designkit-skills",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-05-06T03:21:38.944Z",
      "isPublic": true
    }
  ]
}

Record generated Oct 10, 2026.

Sponsored

Ads related to DesignKit Ecommerce AI Studio and adjacent AI workflows.