← Back to Skills Marketplace
Video pusher
by
SamCheng0717
· GitHub ↗
· v1.0.1
· MIT-0
354
Downloads
1
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install video-pusher
Description
多平台视频/图文内容发布技能集合。支持账号管理、登录状态维护、一键多平台发布。 当用户要求发布内容到抖音、小红书、微信视频号、Threads、Instagram,或管理发布账号时触发。
Usage Guidance
This skill appears to be what it says: local Playwright scripts that open real browser windows to help you upload and fill content, and then require you to manually click the final publish/share button. Before installing: 1) Verify you trust the 'uv' package (it is required to run the provided commands); 2) be aware 'uv run playwright install chromium' will download Chromium (~150MB); 3) the skill will store browser sessions and cookies under video-pusher/profile/ (accounts.json and per-platform profile folders) — review and manage these files carefully (use separate/test accounts if you are concerned); 4) the scripts will be able to read any media file path you supply, so do not pass sensitive file paths; and 5) automated posting may violate some platforms' Terms of Service, so confirm compliance. If you need greater assurance, inspect the 'uv' package source and run the scripts in an isolated environment or on a throwaway account first.
Capability Analysis
Type: OpenClaw Skill
Name: video-pusher
Version: 1.0.1
The video-pusher skill bundle is a legitimate automation tool designed to manage social media accounts and facilitate content publishing across platforms like Douyin, Instagram, and WeChat. It uses Playwright to automate browser interactions, storing session data locally in a 'profile/' directory. The scripts (e.g., vp_accounts.py, publish_douyin.py) are well-structured, lack obfuscation, and include safety features such as requiring manual user confirmation for the final 'Publish' action and providing instructions to exclude sensitive session data from version control via .gitignore.
Capability Assessment
Purpose & Capability
Name/description match the code and instructions: all scripts perform browser automation to publish to the listed platforms and the SKILL.md consistently routes actions to those scripts. The single required binary ('uv') aligns with the documented 'uv run' usage throughout the docs and scripts.
Instruction Scope
SKILL.md and per-subskill SKILL.md files restrict operations to the project's Python scripts and require user confirmation/manual clicks for final publishing. The runtime instructions and code only access the local profile/ directory (accounts.json and per-platform Chromium profiles), the provided media file paths, and the target platform websites. They do not attempt to read other system config or send data to third-party endpoints beyond the social platforms visited during normal publishing.
Install Mechanism
Install requires installing/using the 'uv' tool and running 'playwright install chromium', which downloads Chromium (~150MB). This is expected for Playwright‑based automation, but it is a non-trivial download and relies on the 'uv' package; users should confirm 'uv' is from a trusted source (PyPI) before installing.
Credentials
No environment variables or external credentials are requested. The scripts store/reuse browser sessions and cookies under profile/, which is appropriate for a session‑based automation tool. The skill does not ask for unrelated secrets or access to other credentials.
Persistence & Privilege
The skill persists Chromium session data and accounts.json under a project-local profile/ directory (expected for reuse of logins). It does not request 'always: true' or modify other skills or system-wide agent settings. Autonomous invocation is allowed (platform default) but not coupled with elevated privileges or secret access.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install video-pusher - After installation, invoke the skill by name or use
/video-pusher - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
增加初始化,自动配置安装环境
v1.0.0
Initial release of video-pusher
- Provides a unified skill for managing multi-platform video/text content publishing and account management.
- Supports publishing to Douyin, Xiaohongshu, WeChat Channels, Threads, and Instagram, with login state confirmation before publishing.
- Allows account group creation, platform login/status check, and removal through Python scripts.
- All publishing requires user confirmation and manual submission in browser.
- Includes error guidance for common failures (e.g., not logged in, profile conflicts, structural site changes).
- Enforces usage of absolute file paths and project-root as working directory.
Metadata
Frequently Asked Questions
What is Video pusher?
多平台视频/图文内容发布技能集合。支持账号管理、登录状态维护、一键多平台发布。 当用户要求发布内容到抖音、小红书、微信视频号、Threads、Instagram,或管理发布账号时触发。 It is an AI Agent Skill for Claude Code / OpenClaw, with 354 downloads so far.
How do I install Video pusher?
Run "/install video-pusher" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Video pusher free?
Yes, Video pusher is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Video pusher support?
Video pusher is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux, win32).
Who created Video pusher?
It is built and maintained by SamCheng0717 (@samcheng0717); the current version is v1.0.1.
More Skills