The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill's instructions, scripts, or installation process. The skill performs its described functions using standard audio engineering practices.
[INDIRECT_PROMPT_INJECTION]: The skill processes external MIDI files and JSON structures, which is required for its musical analysis and rendering capabilities.
Ingestion points: midi_inventory.py and render_electronic.py read MIDI files; midi_render.py and midi_transform.py read JSON structure files.
Boundary markers: Not applicable as the skill processes structured binary (MIDI) and text (JSON) data rather than natural language instructions.
Capability inventory: The skill can write files to the designated /mnt/user-data/outputs/ directory and run local Python scripts packaged within its folder.
Sanitization: Input data is parsed using the standard mido and json libraries, providing structural validation.

music-generation