The Agent Skills Directory

[SAFE]: The skill provides legitimate automation scripts for the DaVinci Resolve Studio API, utilizing the official DaVinciResolveScript and Fusion scripting interfaces.
[SAFE]: No suspicious network activity, unauthorized external downloads, or remote code execution patterns were identified in the provided scripts or instructions.
[INDIRECT_PROMPT_INJECTION]: The skill includes a script (gemini.md) to parse FCPXML/EDL files, which are external data sources.
Ingestion points: The audit_fcpxml_timeline function in gemini.md ingests external XML files.
Boundary markers: None explicitly defined for the input data.
Capability inventory: The script is limited to reading the file and printing content to the standard output; it lacks network access or file-write capabilities.
Sanitization: Uses standard xml.etree.ElementTree for parsing. The risk is considered safe due to the limited impact of the operations.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill references standard configuration paths for DaVinci Resolve (e.g., %APPDATA%\Blackmagic Design\... and %PROGRAMDATA%\Blackmagic Design\...). These references are strictly for documentation and environment configuration and do not involve harvesting or exfiltrating sensitive data.

davinci-resolve