The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The script scripts/imiva_mcp.py executes npx -y @infimind/ecom-content-cli@latest. This command fetches and runs code from the npm registry at runtime, which constitutes remote code execution from a source not included in the pre-verified trust list.
[COMMAND_EXECUTION]: The file scripts/imiva_mcp.py uses subprocess.Popen to launch external system commands. While it uses a list for arguments to prevent basic shell injection, it still provides a mechanism for the agent to execute arbitrary external binaries if the parameters are manipulated.
[EXTERNAL_DOWNLOADS]: The skill's configuration and scripts depend on external assets, specifically fetching the @infimind/ecom-content-cli package and communicating with the imiva.ecpro.com API endpoint.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as product materials, selling points, and target audience descriptions (Ingestion points: SKILL.md usage scenarios). This data is interpolated into tool arguments and passed to the imiva_mcp.py script.
Ingestion points: User-provided ecommerce content and product details.
Boundary markers: The script uses JSON-RPC formatting and json.dumps to separate commands from data.
Capability inventory: The skill has the ability to execute shell commands via subprocess and perform network operations.
Sanitization: Employs standard JSON serialization for tool arguments, but lacks explicit content filtering for the generated prompts.

ai-ecommerce-expert-wechat-video-channel-ecommerce-content