← Back to Skills Marketplace
88
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install post-kuaishou-short-videos-infinite
Description
在快手创作者平台发布短视频。触发场景：发布快手视频、上传快手视频、快手发布视频、发快手视频、发布快手短视频。
Usage Guidance
What to consider before installing:
- Understand runtime actions: this skill expects to run GUI automation on your machine (pyautogui + tkinter) and to control an OpenClaw-hosted Chrome instance and the system clipboard and file dialogs. That can interact with other apps and overwrite clipboard contents.
- Ask the author to declare dependencies: the skill metadata should list required binaries/libraries (Python, pyautogui, tkinter) and any required OpenClaw browser availability.
- Prefer interactive invocation: because the agent is instructed to act autonomously on the desktop and "try to resolve popups itself", restrict use to user-invoked runs until you trust it.
- Protect sensitive data: avoid having secrets or other sensitive clipboard contents present when running; provide explicit file paths only when asked.
- Request safer alternatives: prefer using the browser file-input element (upload dialog handled by the browser control) rather than system-level pyautogui clipboard hacks, or a method that doesn't rely on pasting into the OS clipboard.
- Test in a sandbox: run the skill in a non-production environment first to observe behavior and ensure it doesn't mis-click or interact with unexpected UI elements.
If you cannot get these clarifications and guarantees, treat the skill as risky and consider not installing it or only running it under close supervision.
Capability Analysis
Type: OpenClaw Skill
Name: post-kuaishou-short-videos-infinite
Version: 1.0.2
The skill automates video posting on Kuaishou and includes a Python script in SKILL.md that utilizes 'pyautogui' and 'tkinter' to interact with the operating system's file dialog and clipboard. While these high-risk capabilities are plausibly intended to handle file uploads that browser-based tools cannot reach, they grant the agent direct control over the host's keyboard and mouse, which could be misused. No explicit evidence of intentional malice or data exfiltration was found, but the use of OS-level automation qualifies it as suspicious under the provided criteria.
Capability Assessment
Purpose & Capability
The name/description match the instructions (automating posting on Kuaishou). However, the SKILL.md requires running Python GUI automation (pyautogui and tkinter) and controlling an OpenClaw-hosted browser, but the skill metadata does not declare these runtime dependencies or required binaries — a capability/requirement mismatch.
Instruction Scope
The instructions instruct the agent to control the host GUI (file dialogs), manipulate the system clipboard, and autonomously attempt to resolve popups without always asking the user. That grants the agent broad control over the user's desktop environment and clipboard and could cause accidental interactions with other apps or leak clipboard contents; this is beyond a simple API or browser-only automation.
Install Mechanism
There is no install spec (instruction-only), which is lower risk than arbitrary downloads. However, the runtime expects external Python libraries (pyautogui, tkinter) and a hosted Chrome instance; those expectations are not documented in the skill metadata.
Credentials
The skill requests no environment variables or credentials (good). But it requires access to local files and the clipboard and to control the host GUI — these are sensitive capabilities that should be explicitly declared and justified in metadata.
Persistence & Privilege
The skill is not always-enabled, and it does not request system-wide persistence or modifications of other skills. Autonomous invocation is allowed (platform default) but not combined with any declared broad credential access.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install post-kuaishou-short-videos-infinite - After installation, invoke the skill by name or use
/post-kuaishou-short-videos-infinite - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
Version 1.0.2
- No user-visible changes; this is a maintenance or metadata-only update.
- No file changes detected from the previous version.
v1.0.0
Initial release for 快手短视频“一键发布”自动化流程。
- 支持在快手创作者平台完整自动化发布短视频，包括上传、填写描述、封面和声明设置、权限配置等全流程。
- 详细分步指令，涵盖浏览器启动、页面导航、视频上传（含中文路径处理）、描述和标签输入、PK封面设置、AI声明选择、下载权限控制。
- 严格要求全程快照获取元素 ref，杜绝硬编码，保障稳定性。
- 自动识别并确认标签下拉框，避免标签超限造成发布失败。
- 内含 pyautogui 与 tkinter 跨操作系统解决文件对话框中文路径粘贴失效问题。
- 明确列举常见失败原因和关键技术要点，有效防止操作失误。
Metadata
Frequently Asked Questions
What is 发布快手短视频_无限?
在快手创作者平台发布短视频。触发场景：发布快手视频、上传快手视频、快手发布视频、发快手视频、发布快手短视频。 It is an AI Agent Skill for Claude Code / OpenClaw, with 88 downloads so far.
How do I install 发布快手短视频_无限?
Run "/install post-kuaishou-short-videos-infinite" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 发布快手短视频_无限 free?
Yes, 发布快手短视频_无限 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 发布快手短视频_无限 support?
发布快手短视频_无限 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 发布快手短视频_无限?
It is built and maintained by 未来X无限 (@infiniteask); the current version is v1.0.2.
More Skills