ai-ecommerce-expert-ugc-seeding-video

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The helper script scripts/imiva_mcp.py uses the Python subprocess module to bridge the agent with the official IMIVA Model Context Protocol (MCP) tool. The execution is strictly limited to invoking the vendor's command-line interface to facilitate task processing.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx to fetch and run the official @infimind/ecom-content-cli package from the public registry. This is a legitimate and expected behavior for interacting with the vendor's content generation platform.
  • [SAFE]: The skill demonstrates safe credential handling practices by explicitly instructing users to store sensitive authentication tokens in environment variables or client secret areas, rather than hardcoding them within the skill or configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 08:01 AM
Security Audit — agent-trust-hub — ai-ecommerce-expert-ugc-seeding-video