The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses ffmpeg, ffprobe, and python3 to perform audio processing, metadata extraction, and visualisation tasks within a local session directory.- [COMMAND_EXECUTION]: Local Python scripts are dynamically generated via heredoc templates to produce spectrograms using the librosa and matplotlib libraries.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests user feedback to iteratively adjust EQ parameters and filter chains.
Ingestion points: User feedback in the tuning loop and preset files referenced in $ARGUMENTS.
Boundary markers: No explicit boundary markers or instructions are provided to the agent to treat user-provided feedback as untrusted data.
Capability inventory: File system manipulation (mkdir, cp, mv, rm), audio rendering via ffmpeg, and script execution via python3.
Sanitization: No sanitization logic is defined for feedback strings before they are used to influence the agent's logic or parameter generation.

tune-preset