The Agent Skills Directory

[SAFE]: No security issues detected. The skill consists entirely of instructional text and references for copywriting.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local project files (e.g., PRODUCT.md, README.md) to generate copy. This is a standard functional requirement for the skill. The risk is minimized as the skill has no dangerous capabilities such as file modification or network access, and it includes explicit boundaries preventing autonomous actions like publishing content.\n- [DATA_EXFILTRATION]: No hardcoded credentials, sensitive file paths, or unauthorized network operations were identified. The instructions focus on content generation rather than data manipulation.\n- [REMOTE_CODE_EXECUTION]: No patterns of remote script downloading or execution were found. All references are to internal documentation or well-known public repositories for attribution.

suede-copy