The Agent Skills Directory

[SAFE]: The skill focuses on content moderation and UI text improvement. It defines a set of heuristics and checklists to ensure that user-visible strings are professional, clear, and free of sensitive internal implementation details.
[DATA_EXPOSURE]: The skill explicitly instructs the agent to identify 'leaked internals' (e.g., I01-I08 in banned-patterns.md) including API paths, database schema keys, and stack traces that might have been accidentally rendered in the UI. This process is intended to sanitize the product and prevent information disclosure.
[EXTERNAL_DOWNLOADS]: The skill includes a research source file referencing numerous high-reputation domains and organizations such as GOV.UK, NN/group, Intuit, and Elastic. These are used strictly for documentation and research purposes and do not involve automated script execution or package installation.

unslop-copywriting