caption-decorations

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a technical framework for authoring caption decorations in a specific graphical format. No malicious instructions, obfuscation, or data exfiltration patterns were detected.
  • [DYNAMIC_EXECUTION]: The skill employs a domain-specific expression language ($eval) within its JSON definitions to handle measured geometry (e.g., boxW, boxH). These calculations are strictly scoped to graphical properties and do not facilitate arbitrary command or code execution.
  • [COMMAND_EXECUTION]: The skill utilizes specialized MCP tools (picsart_scene_validate and picsart_scene_contact_sheet) for rendering and validation tasks. These tools are consistent with the skill's stated purpose and originate from the recognized author.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 02:44 PM