← Back to Skills Marketplace
🔌

Meitu Beauty

by Meitu · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
320
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install meitu-beauty
Description
对人像照片进行 AI 美颜处理(磨皮、美白、精修五官)。当用户提到美颜、磨皮、美白、精修、beautify、beauty enhance、让照片更好看时触发。仅支持单人照片。
Usage Guidance
This skill appears to do what it says: it calls the meitu CLI and requires your Meitu AK/SK, and will read/write files in ~/.openclaw/workspace/visual/ and ~/.meitu/credentials.json. Before installing, consider: (1) you will need to install a third‑party npm CLI (meitu-cli) and provide API credentials; (2) images processed will be uploaded to Meitu's service via that CLI — avoid sending sensitive/private photos if you do not want them transmitted; (3) confirm you trust the meitu-cli package source (npm) and the Meitu service; (4) if you want the agent to pre-validate faces locally, ask the skill author to clarify whether local face-detection is used or if validation relies solely on Meitu's responses. No regex-scan warnings were present, and nothing in the skill requests disproportionate credentials or unrelated system access.
Capability Analysis
Type: OpenClaw Skill Name: meitu-beauty Version: 1.0.0 The meitu-beauty skill is a legitimate wrapper for the Meitu CLI tool, designed to perform AI-based image enhancement on portraits. It requires standard API credentials and file access permissions consistent with its stated purpose, and the instructions in SKILL.md focus entirely on image validation, CLI execution, and error handling without any signs of malicious intent or prompt injection.
Capability Assessment
Purpose & Capability
Name/description (AI beautify single-person portrait) match the declared requirements: a 'meitu' CLI binary and Meitu AK/SK. Declared file reads/writes (~/.meitu/credentials.json and the visual workspace) align with storing credentials and output images.
Instruction Scope
SKILL.md stays focused on: validating a single-person photo, calling `meitu image-beauty-enhance`, handling outputs and retries. It requires downloading/saving user-provided images (local path, URL, or uploaded image) and will upload them to the Meitu service via the CLI — this is expected but important for privacy. One minor ambiguity: the doc uses '目视检查' (visual check) for face detection; it's not explicit whether the agent uses a local face-detector or relies solely on the meitu CLI. That affects whether local face-detection code would run, but nothing in the skill claims to read unrelated files.
Install Mechanism
Instruction-only skill (no install spec). It suggests installing meitu-cli via `npm install -g meitu-cli`, which is a reasonable, proportional dependency for wrapping that CLI. No downloads from arbitrary URLs or archive extraction are declared.
Credentials
Only MEITU_OPENAPI_ACCESS_KEY and MEITU_OPENAPI_SECRET_KEY are required (with AK as primary). These directly map to the Meitu API use described. The skill also declares reading ~/.meitu/credentials.json as an alternate credential source — consistent with the credential setup instructions.
Persistence & Privilege
always:false and no system-wide modifications. The skill reads/writes only its declared workspace path and the credentials file; it does not request elevated system privileges or modify other skills. Autonomous invocation is allowed (platform default) but not combined with other high-risk factors.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install meitu-beauty
  3. After installation, invoke the skill by name or use /meitu-beauty
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Version 1.1.0 introduces a comprehensive AI-powered portrait beautification workflow, supporting single-person photos only. - Added detailed workflow for input validation, preflight checks, command execution, and output delivery. - Strong input checks: only allows images with exactly one clear face, alerts users on errors or quality issues. - Two beautify strength modes: natural and enhanced, mapped to user intent. - Improved error handling and fallback strategies for common CLI errors and user input problems. - Automatic output directory resolution and standardized output file naming. - Clear dependency and credential requirements for setup.
Metadata
Slug meitu-beauty
Version 1.0.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is Meitu Beauty?

对人像照片进行 AI 美颜处理(磨皮、美白、精修五官)。当用户提到美颜、磨皮、美白、精修、beautify、beauty enhance、让照片更好看时触发。仅支持单人照片。 It is an AI Agent Skill for Claude Code / OpenClaw, with 320 downloads so far.

How do I install Meitu Beauty?

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

Is Meitu Beauty free?

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

Which platforms does Meitu Beauty support?

Meitu Beauty is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Meitu Beauty?

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

💬 Comments