The Agent Skills Directory

[PROMPT_INJECTION]: The skill's instructions are focused on providing a specific service. There are no attempts to override agent safety guidelines, extract system prompts, or bypass restrictions.
[DATA_EXFILTRATION]: No patterns of sensitive data access (like credentials or private files) or unauthorized network transmissions were found. The skill uses standard search queries to fulfill its primary function.
[EXTERNAL_DOWNLOADS]: The skill relies on its own internal reference files and general web searches. It does not attempt to download or execute external scripts or packages.
[REMOTE_CODE_EXECUTION]: There are no commands that download and pipe content to a shell, nor is there any usage of eval, exec, or other dynamic execution functions.
[OBFUSCATION]: The content is clear and readable. No hidden characters, base64-encoded strings, or homoglyph attacks were detected.
[COMMAND_EXECUTION]: The skill does not invoke any system-level commands, privilege escalation tools, or persistence mechanisms.
[INDIRECT_PROMPT_INJECTION]: While the skill processes external search data, it does so for data extraction purposes (prices, trends) and does not have capabilities like file writing or script execution that could be exploited by malicious search results.

ecom-product-radar