The Agent Skills Directory

[PROMPT_INJECTION]: The skill monitors the project's assets/texts directory using a FileWatcher and automatically parses markdown files into the project's state. This creates an indirect prompt injection surface where instructions embedded in media metadata or script files could influence agent behavior.
Ingestion points: Markdown files in assets/texts/ processed by PostPlusWorkspaceRuntime.handleExternalMarkdownEdit in postplus_workspace_runtime.mjs.
Boundary markers: Uses HTML comments like  to separate content blocks.
Capability inventory: The skill has access to the file system, network (PostPlus Cloud), and can execute shell commands via yt-dlp and ffmpeg.
Sanitization: Includes basic unescaping and validation of identifiers, but lacks deep sanitization of the natural language content within blocks.
[COMMAND_EXECUTION]: The skill invokes external command-line utilities to perform media-related tasks. Specifically, it uses spawn to run python3 -m yt_dlp for downloading videos and execFileSync to run ffmpeg for generating test media. On macOS, it also uses the open command to display files to the user.
[EXTERNAL_DOWNLOADS]: The skill downloads media assets (video, audio, and images) from vendor-controlled cloud storage and other remote URLs defined in project manifests.

video-batch-runner