The Agent Skills Directory

[SAFE]: The skill does not contain any prompt injection attempts or instructions to bypass safety guidelines. It provides clear operational constraints and self-checklists.
[SAFE]: Data access is limited to reading repository files relevant to engineering standards, which is required for the skill's primary function. No evidence of credential harvesting or unauthorized file access was found.
[SAFE]: No remote code execution patterns, unauthorized external downloads, or network exfiltration attempts were detected. The skill uses local repository data and user input.
[SAFE]: The skill does not use obfuscation techniques. All instructions and templates are in plain text and clearly define the agent's behavior and boundaries.

guardrails-writer