The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides a Python script (transcribe_audio_pipeline.py) that utilizes the subprocess module to execute the ffmpeg CLI tool. This is a standard and expected behavior for the skill's stated purpose of audio normalization.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources, such as audio/video files and transcript JSON/SRT files, which represents a potential surface for indirect prompt injection if the processed content contains instructions that the agent might inadvertently follow. • Ingestion points: transcribe_audio_pipeline.py (audio/video files), audit_srt.py (SRT subtitle files), and process_meeting_transcript.py (JSON transcript files). • Boundary markers: No explicit delimiter or instruction-ignoring warnings are used when processing input data. • Capability inventory: The skill has the capability to execute the ffmpeg subprocess (SKILL.md) and perform file read/write operations across all scripts. • Sanitization: No specific sanitization or filtering of the content within the ingested files is performed before processing or inclusion in output reports.

macwhisper