ai-ecommerce-expert-vertical-9-16-short-video

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill downloads and executes the @infimind/ecom-content-cli@latest package from the NPM registry at runtime using the npx command. While this is the official tool for the IMIVA platform, executing unpinned remote code is a common surface for supply chain risks.
  • [COMMAND_EXECUTION]: The helper script scripts/imiva_mcp.py uses subprocess.Popen to call shell commands. This is used to wrap the Node.js-based Model Context Protocol (MCP) tool and facilitate interaction with the agent environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied product information which is interpolated into prompts for video generation.
  • Ingestion points: Product names, selling points, target audience, and descriptions provided via the create_video_generation_task tool.
  • Boundary markers: Absent. User-provided data is not delimited from system instructions.
  • Capability inventory: The skill can trigger remote video generation tasks and budget deductions on the IMIVA platform.
  • Sanitization: No filtering or escaping is applied to user-provided strings before they are used in the generated prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 08:02 AM
Security Audit — agent-trust-hub — ai-ecommerce-expert-vertical-9-16-short-video