The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown instructions and metadata without any executable scripts or binary files. There are no signs of malicious patterns such as credential harvesting, exfiltration, or persistence.
[EXTERNAL_DOWNLOADS]: A reference to an external GitHub repository is provided for attribution of the original work. This is a static link to source material and does not perform any runtime downloads or remote code execution.
[INDIRECT_PROMPT_INJECTION]: The skill defines a process for the agent to edit user-provided content such as READMEs and tickets. While this data is untrusted, the skill does not use any capabilities (file system access, network operations, or tool calls) that could be leveraged by an attacker through the processed text.

unslop