The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill relies on npx -y @infimind/ecom-content-cli@latest to download and execute an external package from the NPM registry whenever the helper script is invoked. This pattern executes code from a remote source at runtime.
[COMMAND_EXECUTION]: The file scripts/imiva_mcp.py uses subprocess.Popen to launch the npx command. This establishes a bridge between the Python environment and shell-based Node.js tools, creating a surface for command execution.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data through the create_koc_collage_task tool, which accepts parameters like productName, sellingPoints, and targetAudience.
Ingestion points: User-provided strings for product details in SKILL.md (e.g., productName, sellingPoints).
Boundary markers: None identified; parameters are passed directly to the MCP server as JSON fields.
Capability inventory: The skill has the capability to generate content and perform network operations via the IMIVA MCP server.
Sanitization: No input sanitization or validation logic is present in the imiva_mcp.py wrapper script before passing data to the subprocess.

ai-ecommerce-expert-influencer-ugc-product-image