skills/wubin1836/ai-ecommerce-expert-agent-skills/ai-ecommerce-expert-shopify-ecommerce-content/Gen Agent Trust Hub
ai-ecommerce-expert-shopify-ecommerce-content
Warn
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill relies on
npx -y @infimind/ecom-content-cli@latestas documented inSKILL.mdand implemented inscripts/imiva_mcp.py. This pattern downloads and executes external code from the NPM registry at runtime without fixed versioning or integrity verification, which could allow for supply-chain attacks. - [DYNAMIC_EXECUTION]: The Python script
scripts/imiva_mcp.pyusessubprocess.Popento execute the Node.js package. This mechanism facilitates the execution of external scripts within the host environment, a risk factor identified by static analysis. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes user-supplied text inputs to generate marketing content. Ingestion points: product details, style preferences, and additional description fields in
SKILL.md. Boundary markers: None identified; the skill lacks structural delimiters to isolate instructions from data. Capability inventory: Local file system read access for material processing and network communication with the IMIVA API (imiva.ecpro.com). Sanitization: The wrapper scriptscripts/imiva_mcp.pyperforms no validation or escaping of the user-provided arguments before processing.
Audit Metadata