← Back to Skills Marketplace
jkin8010

MoPNG 图片工具

by jkin8010 · GitHub ↗ · v0.1.7 · MIT-0
cross-platform ✓ Security Clean
235
Downloads
0
Stars
1
Active Installs
8
Versions
Install in OpenClaw
/install mopng-api
Description
使用 mopng.cn (MoPNG) API 进行图片处理,包括智能抠图、高清放大、智能扩图、图片翻译、文生图、图生图等功能。支持 API Key 鉴权;处理本地文件时会上传至 MoPNG 服务。
Usage Guidance
This skill appears to do what it says: it will upload local images to mopng.cn and call the MoPNG API using MOPNG_API_KEY. Before installing, keep these points in mind: - Treat MOPNG_API_KEY like any API secret: set it in the host's private env/config, do not paste it into chat, and rotate it if leaked. - Local images will be uploaded to the MoPNG service; do not process sensitive images unless you trust the service and accept that transmission/storage occurs off-host. - Set OPENCLAW_WORKSPACE to a restrictive absolute path so the script can only read/write within that directory. - Only add domains you trust to MOPNG_EXTRA_ALLOWED_IMAGE_HOSTS; allowing arbitrary domains can defeat the remote-URL guardrails. - The README points to GitHub/Gitee locations; if you require stronger assurance, review the full repository history at those remote sources (or run the included tests and bandit locally) before enabling the skill in production.
Capability Analysis
Type: OpenClaw Skill Name: mopng-api Version: 0.1.7 The mopng-api skill provides image processing features such as background removal, upscaling, and image generation via the mopng.cn API. The implementation in scripts/mopng_api.py includes robust security controls, including SSRF mitigation for remote URLs, path traversal protection to ensure file operations remain within the designated workspace, and strict validation of image headers and dimensions. All network communication is restricted to the official API domain (mo-api.mopng.cn), and the skill follows the principle of least privilege by using only the Python standard library.
Capability Assessment
Purpose & Capability
Name/description, required env vars (MOPNG_API_KEY, OPENCLAW_WORKSPACE), and the included Python client all align with an image-processing API integration. The declared binary requirements (python3 and 'uv') are reasonable: python3 is needed to run the provided script; 'uv' is referenced in README/test/dev flows (used for local sync/test commands) and is not an unexplained credential or cloud-level secret.
Instruction Scope
SKILL.md describes only operations necessary for the API (uploading local images to MoPNG, providing remote URLs under an allowlist, running text-to-image, etc.). It explicitly warns that local images are uploaded, restricts allowed remote hosts, enforces workspace path constraints, and does not instruct reading unrelated host files or unrelated credentials. The instructions to run pytest and bandit are dev/test steps and not part of runtime behavior.
Install Mechanism
No install spec is provided (instruction-only skill) and the runtime uses a single Python script that depends only on the standard library. There are no downloads from untrusted URLs or archive extraction steps in the manifest.
Credentials
Requested environment variables map to the function: an API key for MoPNG and an optional workspace root to constrain file access. No unrelated credentials or high-privilege secrets are requested. One minor note: MOPNG_EXTRA_ALLOWED_IMAGE_HOSTS can expand which remote hosts are permitted — that is user-controlled and should only be set to trusted hosts.
Persistence & Privilege
The skill is not forced always-on (always:false) and uses standard autonomous invocation defaults. It does not request to modify other skills or system-wide configs. No elevated persistence or cross-skill privileges are requested.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install mopng-api
  3. After installation, invoke the skill by name or use /mopng-api
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.7
No file changes detected for version 0.1.7. - No updates or modifications in this release. - Functionality and documentation remain unchanged.
v0.1.6
- 增加 OPENCLAW_WORKSPACE、MOPNG_EXTRA_ALLOWED_IMAGE_HOSTS 环境变量说明,完善多端和安全运行场景支持 - 强化安全约束:明确本地图片处理将上传服务端,限制远程 URL 主机与协议、提示词长度、新增敏感词过滤开关说明 - 详细补充环境变量注入、私密配置、密钥安全 Best Practice 指引 - 优化输出路径规范与路径校验细节,明确工作区机制 - 更新文生图、图生图命令参数说明,增加提示词字符上限与敏感词过滤参数说明 - 新增 scripts/ 目录下代码自动进行 pytest/bandit 扫描的仓库合规说明
v0.1.5
- 文档结构精简,命令使用说明更加简明直观 - 移除详细安装与环境设置流程,API Key 配置介绍更简短 - 分组并高亮各项功能的基本用法与常用选项 - 统一例子为 Claude 聊天命令格式,便于用户直接复制使用 - 进一步明确功能限制及异步任务说明
v0.1.4
- 更新了 API 密钥申请和环境变量设置说明,加入了 Claude Code 配置方法与示例测试命令 - `--output` 参数说明更加明确,写法为相对路径时会自动存入 `outputs/mopng-api/` 目录 - 新增 `.gitignore` 和 `uv.lock` 文件以支持项目管理 - 优化文档结构和使用流程说明,便于新用户快速上手
v0.1.3
- 更新 API 说明和文档链接,从 qise.cc 变更为 mopng.cn - 配置说明新增对「Claude Code」的环境变量设置方式说明 - 命令行和功能参数保持不变 - 优化环境变量设置步骤及帮助信息,便于多平台兼容与使用
v0.1.2
- No changes detected in this version; content remains the same as the previous release.
v0.1.1
fix: API response field compatibility and async task polling
v0.1.0
- Initial release of mopng-api skill. - Provides image processing functions using qise.cc (MoPNG) API, including background removal, upscaling, outpainting, image translation, text-to-image, and image-to-image. - Supports API Key authentication via environment variable `MOPNG_API_KEY`. - Command-line interface for each feature with detailed options. - Enforces input/output file security checks and size restrictions. - Adds async support for longer tasks and result polling. - Includes model discovery and helpful usage examples.
Metadata
Slug mopng-api
Version 0.1.7
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 8
Frequently Asked Questions

What is MoPNG 图片工具?

使用 mopng.cn (MoPNG) API 进行图片处理,包括智能抠图、高清放大、智能扩图、图片翻译、文生图、图生图等功能。支持 API Key 鉴权;处理本地文件时会上传至 MoPNG 服务。 It is an AI Agent Skill for Claude Code / OpenClaw, with 235 downloads so far.

How do I install MoPNG 图片工具?

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

Is MoPNG 图片工具 free?

Yes, MoPNG 图片工具 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does MoPNG 图片工具 support?

MoPNG 图片工具 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created MoPNG 图片工具?

It is built and maintained by jkin8010 (@jkin8010); the current version is v0.1.7.

💬 Comments