The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external video and audio files provided by the user via the polysync command-line tool. This ingestion of untrusted data combined with shell execution capabilities represents a standard attack surface.
Ingestion points: Media files defined as REFERENCE and SOURCE are passed to the tool for analysis in SKILL.md.
Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore potentially malicious content embedded in the media file metadata or filenames.
Capability inventory: The skill includes shell execution capabilities via python3 -m pip install and the polysync CLI.
Sanitization: There is no explicit sanitization step for the user-provided filenames before they are interpolated into the bash command strings.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands to manage dependencies and perform the synchronization logic. Specifically, it uses python3 -m pip install -U polysync and polysync sync commands.
[EXTERNAL_DOWNLOADS]: The skill fetches the polysync package from the Python Package Index (PyPI) to support its functionality, as documented in the setup instructions.

wjs-syncing-multicam