upscale-and-interpolate
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists exclusively of markdown documentation and instructions for the agent; it does not include any scripts, binary files, or installation commands.- [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection by instructing the agent to read and follow parameters from external files such as the creative brief.
- Ingestion points: brief/creative-brief.md, brief/tools-and-models.md (referenced in SKILL.md).
- Boundary markers: None. The skill does not instruct the agent to use specific delimiters or to ignore potential instructions within the ingested data.
- Capability inventory: The workflow involves reading local files, writing processed videos to local directories, and invoking external command-line tools for video processing.
- Sanitization: None specified.- [SAFE]: No malicious patterns, obfuscation, or unauthorized exfiltration techniques were detected. The skill operates on local project files and uses established industry software names (e.g., Topaz, Real-ESRGAN).
Audit Metadata