ai-video-editing-software
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to communicate with the vendor's API at mega-api-prod.nemovideo.ai using standard HTTP requests for video processing, credits management, and rendering. These operations are required for the skill's advertised functionality.
- [SAFE]: Authentication is handled via a dedicated environment variable (NEMO_TOKEN) and local configuration path (~/.config/nemovideo/), which is consistent with secure development practices for CLI-based tools.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection as it processes raw media footage and user-provided captions. While this allows for the ingestion of untrusted data, the skill does not grant the agent capabilities that could be maliciously leveraged through this vector.
- Ingestion points: Raw footage metadata, transcription requests, and user-provided edit instructions in SKILL.md.
- Boundary markers: None explicitly defined for media processing results.
- Capability inventory: Network POST/GET requests for media management and rendering in SKILL.md.
- Sanitization: None detected.
Audit Metadata