The Agent Skills Directory

[SAFE]: Comprehensive analysis across all ten threat categories revealed no security issues. The skill's architecture and use of tools align with its stated purpose of product research.
[COMMAND_EXECUTION]: The skill utilizes the sessions_spawn tool to manage parallel sub-agents for specialized research tasks. This is a standard and safe implementation of a multi-agent workflow on supported AI platforms.
[EXTERNAL_DOWNLOADS]: The skill performs web searches and page fetches to retrieve product information from trusted retail sources such as Amazon.sa, Jarir, and noon.com. These activities are limited to public data gathering for shopping research.
[PROMPT_INJECTION]: The sub-agent prompts include clear role definitions and constraints, reducing the risk of accidental behavior override. No instructions aimed at bypassing safety filters or extracting system prompts were found.

personal-shopper