The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional guidelines for writing and reviewing documentation. It does not contain executable scripts, network operations, or sensitive data access patterns.
[NO_CODE]: The content is purely markdown-based instructions and templates for README files. While it provides examples of commands like bun add or bun test, these are intended as documentation for users and are not executed by the skill itself.
[INDIRECT_PROMPT_INJECTION]: The skill involves reading and writing README files which are external data sources. However, the instructions focus on structural and content conventions (such as keeping license information and responsibility sections) rather than unsafe interpolation or automated execution of content found within those files. The risk is consistent with standard agent text-processing tasks.

kstack-readme