The Agent Skills Directory

[SAFE]: The skill documents the use of the jq utility for processing JSON data, which is a standard development task.
[SAFE]: The 'Security Guardrails' section provides explicit defensive instructions to the agent, such as requiring defensive quoting to prevent shell injection and restricting the extraction of sensitive fields from configuration files.
[SAFE]: The included SAFETY.md file provides a comprehensive safety evaluation with test cases demonstrating how the guardrails mitigate various risks, including data exfiltration and social engineering attempts.
[SAFE]: External links are limited to the official jq website and the author's platform (faberlens.ai), with no suspicious or unauthorized remote code execution patterns detected.

jq-hardened