The Agent Skills Directory

[SAFE]: The skill serves as a thin alias that redirects the agent to use the brainstorm skill. It does not perform any sensitive operations or external communication directly.
[NO_CODE]: The skill consists entirely of markdown instructions and metadata, with no executable scripts or binary dependencies.
[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface through the interpolation of user-supplied arguments into the prompt context.
Ingestion points: User input provided via the $ARGUMENTS variable in SKILL.md.
Boundary markers: Input is encapsulated within <feature_description> tags to provide structural separation.
Capability inventory: The skill instructs the agent to read a local instruction file (../brainstorm/SKILL.md) and interact with GitHub issues.
Sanitization: No explicit sanitization or escaping of the interpolated string is performed.

req