The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to invoke a Python script (/sandbox/.hermes-data/workspace/omni-video-analyze.py) using the terminal tool. This is the core functionality for processing media files and generating summaries or transcripts.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes untrusted user-provided media files and directory manifests (chunks.json, manifest.json).
Ingestion points: User-specified file paths for videos, audio, images, and PDF directories, as well as associated JSON manifest files in *-chunks/ or *-longvideo/ directories.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded content in the processed data.
Capability inventory: The skill has the capability to execute shell commands via the terminal tool.
Sanitization: The instructions do not mention any sanitization or validation of the input file content beyond basic file type handling.

video-analyze