The Agent Skills Directory

[SAFE]: The skill is composed entirely of natural language instructions and lacks any executable scripts, external dependencies, or network-enabled tools.
[INDIRECT_PROMPT_INJECTION]: Although the skill is designed to process external technical data (such as code and issue logs), it does not define or request access to any tools that could be exploited for data exfiltration or system modification. The absence of dangerous capabilities (file-write, network, command execution) renders this attack surface non-exploitable in this context.

to-prompt