Skills
Navigate
skills/wubin1836/ai-ecommerce-expert-agent-skills/ai-ecommerce-expert-koc-3x3-grid/Gen Agent Trust Hub

ai-ecommerce-expert-koc-3x3-grid

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The script scripts/imiva_mcp.py dynamically downloads and executes the @infimind/ecom-content-cli@latest package using npx at runtime. This pattern introduces a dependency on the availability and integrity of the remote NPM registry and the vendor's unversioned package.
  • [COMMAND_EXECUTION]: The skill performs shell command execution using subprocess.Popen in scripts/imiva_mcp.py to bridge the agent's requests to the IMIVA MCP CLI tool.
  • [PROMPT_INJECTION]: The skill processes untrusted user inputs that are passed to external tools, creating a surface for indirect prompt injection.
  • Ingestion points: Arguments for the create_koc_collage_task tool, such as productName and sellingPoints, which are processed in scripts/imiva_mcp.py.
  • Boundary markers: The bridge script does not implement explicit boundary markers or delimiters to isolate user data from tool instructions.
  • Capability inventory: The skill has capabilities for shell command execution (subprocess.Popen) and network requests (via the MCP tool) in scripts/imiva_mcp.py.
  • Sanitization: There is no evidence of sanitization or validation of user-provided strings before they are forwarded to the underlying MCP package.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 08:01 AM
Security Audit — agent-trust-hub — ai-ecommerce-expert-koc-3x3-grid