The Agent Skills Directory

[SAFE]: The skill provides instructions for a conversion copywriter role, focusing on marketing principles like clarity, benefit-driven copy, and specificity. No malicious patterns such as prompt injection, data exfiltration, or unauthorized command execution were found.
[DATA_EXPOSURE_&_EXFILTRATION]: The skill requests to read a local file .claude/product-marketing-context.md to gather context. This is a standard practice for personalizing agent responses and does not involve exfiltrating data to external servers or accessing sensitive system credentials.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests data from a local context file, it lacks capabilities to execute commands, perform network operations, or write to the file system. The risk of indirect prompt injection is negligible as the processed data is only used to generate text output for the user.

copywriting