video-editing
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: Provides standardized FFmpeg utility commands for video manipulation tasks such as trimming, concatenating, and reframing. These commands are localized to the provided media assets and follow standard usage patterns.
- [EXTERNAL_DOWNLOADS]: Facilitates the generation of media assets (voiceovers, music, and images) through requests to well-known service providers including ElevenLabs and fal.ai. These interactions use established official API endpoints.
- [SAFE]: Employs secure development practices by instructing the use of environment variables for API key storage in Python scripts. The skill also describes a workflow for processing media transcripts to generate automation commands. Evidence for data processing surface: Ingestion points: raw.mp4 and transcripts (SKILL.md). Boundary markers: Absent. Capability inventory: FFmpeg subprocess calls and network requests (SKILL.md). Sanitization: Absent. This represents a functional surface inherent to the video editing use-case.
- [REMOTE_CODE_EXECUTION]: The skill uses npx remotion render and LLM-generated FFmpeg scripts to automate video composition. While this involves executing generated commands, it is the primary intended function of the workflow and uses standard media processing tools.
Audit Metadata