The Agent Skills Directory

[SAFE]: The skill serves as a directory and router for other growth-oriented skills. It provides a structured decision-making process to select the most appropriate specialist based on user input.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to recommend installation commands for missing sub-skills using npx skills add. This is a standard mechanism for platform extension and requires user authorization to execute.
[EXTERNAL_DOWNLOADS]: References are made to the author's own domain gingiris.tools for additional resources and services. This is documented clearly as a vendor resource and does not involve automated or hidden downloads.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill includes explicit instructions to avoid requesting secret values in chat and to use secure environment variables. It also mandates keeping customer data out of public artifacts.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user requests to determine routing, it does not ingest untrusted external data (like web scrapings) in its own context. The risk of indirect injection is low as it primarily acts as a classification layer.
[PRIVILEGE_ESCALATION]: There are no requests for elevated permissions or unsafe system modifications. It emphasizes getting explicit approval for consequential actions like publishing or outreach.

gr