← Back to Skills Marketplace
108
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install meitu-carousel
Description
一键生成轮播套组,封面+内页风格统一。适用于小红书组图笔记、知识卡片轮播、产品介绍套图。当用户提到套组、组图、轮播图、轮播套组、知识卡片套图、产品套图时触发。
Usage Guidance
This skill appears to do what it says: it runs the Meitu CLI and needs your Meitu API keys. Before enabling/installing, consider: 1) Only provide MEITU_OPENAPI_ACCESS_KEY/SECRET if you trust the Meitu service and the meitu-cli package source; prefer creating API keys with limited scope. 2) Installing meitu-cli via npm -g modifies your system packages; review the npm package (meitu-cli) origin before installing. 3) The skill will read from and write to your OpenClaw visual workspace (~/.openclaw/workspace/visual/) and may create memory/observation files; back up any important project files if you’re concerned. 4) If you keep credentials in ~/.meitu/credentials.json, be aware the skill may read that file as an alternative to env vars. If any of these behaviors are unacceptable, do not provide the credentials or do not enable project-mode file access.
Capability Analysis
Type: OpenClaw Skill
Name: meitu-carousel
Version: 1.0.0
The skill bundle is a legitimate tool for generating social media carousel images using the Meitu AI platform's CLI (`meitu-cli`). It requires standard API credentials (stored in `~/.meitu/credentials.json`) and file system access to a dedicated workspace (`~/.openclaw/workspace/visual/`) to manage design assets and user preferences. The instructions in `SKILL.md` and the supporting `references/` files are highly detailed, focusing on design quality, typography, and a 'Memory Protocol' for persisting user stylistic choices. No evidence of data exfiltration, malicious command execution, or harmful prompt injection was found; the requested permissions are strictly aligned with the stated functionality.
Capability Assessment
Purpose & Capability
Name/description (生成轮播套组) match the declared requirements: a meitu CLI binary and Meitu API keys. The SKILL.md only calls meitu image-poster-generate and related auth/verify commands which are necessary for an image-generation CLI integration.
Instruction Scope
Instructions are detailed and largely confined to generating cover and inner pages via the meitu CLI and optionally reading/writing project memory under the OpenClaw visual workspace. The skill will read openclaw.yaml, DESIGN.md, and various $VISUAL memory files when in project mode; it will create/write files under the visual workspace (e.g., output_dir, memory files). This is coherent with a project-mode assistant but users should be aware the skill will read and persist non-sensitive project-specific files in ~/.openclaw/workspace/visual/ when enabled.
Install Mechanism
This is an instruction-only skill (no install spec). It recommends installing meitu-cli via npm (npm install -g meitu-cli), which is a standard package manager flow and appropriate for a CLI-based integration. No remote arbitrary downloads or extract-from-URL steps are present in the skill bundle.
Credentials
The only required environment variables are MEITU_OPENAPI_ACCESS_KEY and MEITU_OPENAPI_SECRET_KEY (with file fallback ~/.meitu/credentials.json), which directly correspond to the declared external service (Meitu). No unrelated secrets or credentials are requested.
Persistence & Privilege
always:false (normal). The skill requests read/write access to the user's OpenClaw visual workspace (~/.openclaw/workspace/visual/) and may create/append memory/observation files per the memory-protocol. This is appropriate for a project-mode feature but means the skill will persist user preferences and generated outputs to disk — users should be comfortable with those files being created/modified.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install meitu-carousel - After installation, invoke the skill by name or use
/meitu-carousel - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
**meitu-carousel 1.1.0 – Initial Release**
- 全新上线:一键生成风格统一的轮播套组(封面+多内页),支持小红书组图、知识卡片、产品介绍等场景
- 明确依赖 meitu-cli 工具与美图开放平台 API 凭证,详细列出安装、配置及目录要求
- 支持自动拆分与生成封面文案与多页内页文案,严格规范标题、正文字体、排版与内容生成规则
- 封面/内页配色风格自动统一,并支持项目模式下个性化约束与偏好读取
- 允许反复细化(最多 3 轮)设计风格、配色、排版和文案,配合用户需求灵活调整
- 输出按日期和页序自动命名,适合视觉项目归档与版本管理
Metadata
Frequently Asked Questions
What is Meitu Carousel?
一键生成轮播套组,封面+内页风格统一。适用于小红书组图笔记、知识卡片轮播、产品介绍套图。当用户提到套组、组图、轮播图、轮播套组、知识卡片套图、产品套图时触发。 It is an AI Agent Skill for Claude Code / OpenClaw, with 108 downloads so far.
How do I install Meitu Carousel?
Run "/install meitu-carousel" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Meitu Carousel free?
Yes, Meitu Carousel is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Meitu Carousel support?
Meitu Carousel is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Meitu Carousel?
It is built and maintained by Meitu (@meitu); the current version is v1.0.0.
More Skills