← Back to Skills Marketplace
qianchen94

Skin Pro

by qianchen94 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
33
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install skin-pro
Description
人脸皮肤分析专业版。调用 RageHealth 开放接口 /face/skin-pro,对人脸图片进行 28+ 项专业皮肤检测,包括水油平衡、毛孔粗大、色斑、敏感程度、皱纹细纹、黑头、痘痘、黑眼圈、眼袋、肤质、肌龄、综合评分等。当用户上传人脸照片要求"皮肤分析"、"测肤质"、"看皱纹/毛孔/痘痘/色斑"、"测肌龄...
README (SKILL.md)

皮肤分析专业版(skin-pro)

何时使用

  • 用户上传人脸正面照片,要求专业皮肤检测
  • 用户询问皮肤年龄、肤质、毛孔、痘痘、色斑、皱纹、敏感、黑头、黑眼圈、眼袋等
  • 用户要求皮肤综合评分或多区域细分检测

接口元数据

  • 请求方式POST
  • URLhttps://facepro.ragehealth.cn/openapi-test/face/skin-pro
  • Content-Typemultipart/form-data
  • 认证头AccessKeySignature(每次调用前需重新生成)

请求参数

参数 类型 必填 说明
imageUrl String 二选一 图片 URL
imageFile File 二选一 图片二进制(jpg/png/webp,≥640×640,≤5MB;优先级高于 imageUrl
faceIdDetect Boolean 是否开启 faceId 检测
userGroup String 条件必填 faceIdDetect=true 时必填,作为人脸 id 隔离分组
analyseStrategy Int 计分策略:1=skin-pro 默认(默认值),2=察颜小程序计分

调用方式

python scripts/call_skin_pro.py \
  --image-url https://example.com/face.jpg \
  [--image-file ./face.jpg] \
  [--face-id-detect --user-group \x3Cgroup>] \
  [--analyse-strategy 1|2] \
  [--output result.json]

凭证由脚本自动从环境变量 SKIN_PRO_AK / SKIN_PRO_SK 读取,不要作为参数传入。首次使用前需前往 \x3Chttps://chayan-test.ragehealth.cn/client> 注册申请 AK/SK,写入 scripts/.env。脚本内部会生成 Signature 并以 multipart/form-data 提交。

执行步骤

  1. 校验输入:必须有 imageUrl 或本地图片;图片需 jpg/png/webp、≥640×640、≤5MB。
  2. 调脚本:拿到 JSON;success=false 时提示用户重拍或检查图片质量。
  3. 质量门禁:先看 data.image_quality,命中以下任一项 → 不出报告,提示重拍
    • brightness.grade != 'zhengchang'(过亮/过暗)
    • blur > 0.3(模糊)
    • face_ratio \x3C 0.5(脸太小,官方默认阈值)
    • abs(face_orientation.yaw) > 15abs(pitch) > 15(侧脸/抬低头过度)
    • hair_occlusion > 0.3(刘海遮挡严重)
  4. 解读关键指标(从 data 中按需取):
    • 综合:skin_score / skin_age / aging_index / skd / skin_rank
    • 肤质:skin_type / skin_tone
    • 主要问题:acne / acne_mark / blackhead / blackhead_count / enlarged_pore_countdict,按 5 个区域分别计数)/ melanin / sensitivity / oily_intensity
    • 皱纹:forehead_wrinkle / crows_feet / eye_finelines / glabella_wrinkle / nasolabial_fold / wrinkle_count
    • 眼部:eye_pouch / dark_circle
    • 严重度汇总:analyse_resultlist[dict],每项含 type(英文键名)、score(0~100)、label(良好/轻度/中度/重度)、levels;左右脸项目额外含 leftScore/rightScore/leftLabel/rightLabel
    • 可视化:display_img(带标注展示图)/ face_maps(各维度热力图)
  5. 生成报告:综合分 → 肤质 → 主要问题 Top3(优先 label in {中度,重度},其次按 100-score 降序)→ 改善建议 → 附 display_img

详细响应字段说明见 references/response_schema.md

安全 & 合规

⚠️ 必须提醒:

  • 检测结果仅供参考,不能替代专业医美/皮肤科诊断
  • 严重皮肤问题应建议线下面诊
  • 不在响应中泄露 AccessKey / SecretKey

示例

用户:帮我用专业版分析下这张照片的皮肤。imageUrl=https://example.com/face.jpg

操作(凭证已在宿主环境配置完成,工具调用只携带业务参数):

python scripts/call_skin_pro.py \
  --image-url https://example.com/face.jpg \
  --analyse-strategy 1

回复(节选)

  • 综合评分:78 分,肌龄 26.4,肤质:混合型
  • 主要问题:毛孔(中度,左脸颊 220 个)、黑头(轻度,58 个)、色沉(面积 12%)
  • 皱纹:抬头纹 1,鱼尾纹 2/2
  • 建议:加强清洁控油 + 抗氧化,必要时线下面诊
Usage Guidance
Review this skill before installing. Only use it with face photos the user has consented to upload, do not allow arbitrary endpoint values, and avoid passing arbitrary local file paths. The maintainer should pin or allowlist the RageHealth endpoint, add code-level image validation, and declare the required SKIN_PRO_AK/SKIN_PRO_SK credentials in metadata.
Capability Analysis
Type: OpenClaw Skill Name: skin-pro Version: 1.0.0 The 'skin-pro' skill is a legitimate integration for a professional skin analysis service. The core logic in `scripts/call_skin_pro.py` handles image uploads and API authentication (AES-CBC signature generation) to a specific endpoint (facepro.ragehealth.cn) using environment variables for credentials. The instructions in `SKILL.md` are well-structured, focusing on input validation and result interpretation without any signs of malicious intent, data exfiltration, or harmful prompt injection.
Capability Tags
crypto
Capability Assessment
Purpose & Capability
The stated purpose is to analyze a user-provided face photo through RageHealth, and the artifacts disclose that an image URL or image file is submitted to an external API. Face photos and skin-analysis results are sensitive, so users should understand this external processing.
Instruction Scope
SKILL.md documents a fixed RageHealth endpoint and image inputs, but the script exposes an undocumented --endpoint override and does not enforce that --image-file is actually an image before posting it.
Install Mechanism
There is no automatic install step or remote installer. The included Python requirements are ordinary for HTTP requests, cryptographic signing, and optional .env loading.
Credentials
The script requires SKIN_PRO_AK and SKIN_PRO_SK from environment or .env, which is expected for the RageHealth API, but the registry metadata says no required environment variables or primary credential.
Persistence & Privilege
No background worker, self-starting behavior, or local persistent agent state is shown. Optional faceIdDetect may create or return a remote face identifier, but it is documented as optional.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install skin-pro
  3. After installation, invoke the skill by name or use /skin-pro
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
skin-pro 1.0.0 - 首个版本发布,支持对人脸图片进行 28+ 项专业皮肤检测。 - 检测内容涵盖水油平衡、皱纹、色斑、毛孔、黑头、痘痘、敏感、黑眼圈、眼袋、肌龄、肤质及综合评分等。 - 自动校验图片质量,对不合格图片引导用户重拍。 - 检测结果详细分项展示,含综合评分、问题分级、可视化标注图。 - 明确安全须知,结果仅供参考,严重问题建议线下面诊。
Metadata
Slug skin-pro
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Skin Pro?

人脸皮肤分析专业版。调用 RageHealth 开放接口 /face/skin-pro,对人脸图片进行 28+ 项专业皮肤检测,包括水油平衡、毛孔粗大、色斑、敏感程度、皱纹细纹、黑头、痘痘、黑眼圈、眼袋、肤质、肌龄、综合评分等。当用户上传人脸照片要求"皮肤分析"、"测肤质"、"看皱纹/毛孔/痘痘/色斑"、"测肌龄... It is an AI Agent Skill for Claude Code / OpenClaw, with 33 downloads so far.

How do I install Skin Pro?

Run "/install skin-pro" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Skin Pro free?

Yes, Skin Pro is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Skin Pro support?

Skin Pro is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Skin Pro?

It is built and maintained by qianchen94 (@qianchen94); the current version is v1.0.0.

💬 Comments