The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the playwright Python package and the Chromium browser binary via the playwright install command. These are well-known, industry-standard tools for browser automation and are considered safe sources.
[COMMAND_EXECUTION]: The instructions direct the agent to execute a shared local script (../../shared/scripts/web_publisher.py) to perform actions such as logging in, planning, and publishing content. This script is part of the skill's own project structure and represents standard tool usage for its stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill provides an interface to process user-supplied video metadata (titles and descriptions) for publication on WeChat Channels.
Ingestion points: The publish and plan commands in SKILL.md accept user-controlled strings for the --title argument.
Boundary markers: No explicit delimiter or sanitization instructions are provided in the skill documentation to separate user content from the command execution flow.
Capability inventory: The skill possesses the capability to read local video files and transmit data over the network to the WeChat platform.
Sanitization: The skill assumes the underlying web_publisher.py handles sanitization of the provided strings before interacting with the browser DOM.

skill-channels-upload