Skills
skills/benzema216/dreamina-claude-skills/dreamina-foreground-segmentation/Gen Agent Trust Hub

dreamina-foreground-segmentation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • EXTERNAL_DOWNLOADS (LOW): The skill makes network requests to https://jimeng.jianying.com. This domain is not included in the trusted sources list.
  • DATA_EXFILTRATION (LOW): The skill transmits an image_uri and a user-provided sessionid (via cookies) to an external domain. While this is necessary for the stated functionality, it involves the transfer of session-specific data to an external service.
  • INDIRECT_PROMPT_INJECTION (LOW): The skill has an ingestion point via the image_uri parameter.
  • Ingestion points: image_uri in foreground_segmentation function.
  • Boundary markers: None; the URI is interpolated directly into a JSON string within the draft_content object.
  • Capability inventory: requests.post used to send data externally.
  • Sanitization: None; the input is assumed to be a valid URI string.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:39 PM