higgsfield-image-shots
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation and reference guide for cinematography terms (shot framing, angles, and implied movement) optimized for AI image generation models. It contains no scripts, external dependencies, or network-enabled commands.
- [DATA_EXPOSURE_&_EXFILTRATION]: No sensitive data, hardcoded credentials, or network exfiltration patterns were detected. All examples are localized to image prompt generation strings.
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates that interpolate external data (e.g.,
[img 1]reference IDs or character descriptions) into prompts. While this is an expected feature for image generation skills, it represents a theoretical surface where malicious input could influence the final prompt construction. However, there are no capabilities in this skill to execute commands or exfiltrate data. - Ingestion points: Interpolation of
[img 1]and subject descriptions inSKILL.mdprompt templates. - Boundary markers: None present in the templates.
- Capability inventory: None; the skill is documentation-only and contains no subprocess calls, file writes, or network operations.
- Sanitization: None provided within the skill text itself.
- [OBFUSCATION]: No hidden content, encoded strings, or suspicious character substitutions were found.
Audit Metadata