The Agent Skills Directory

[SAFE]: The skill is composed entirely of markdown text and strategy frameworks. There are no executable scripts, binaries, or automated tasks associated with this skill.
[INDIRECT_PROMPT_INJECTION]: The skill interpolates user-supplied data into its strategy prompt via the $ARGUMENTS placeholder. Ingestion points: User input enters the context through the $ARGUMENTS variable in SKILL.md. Boundary markers: The input is highlighted with bold text but lacks explicit isolation delimiters. Capability inventory: The skill provides no tool access, file system permissions, or network capabilities. Sanitization: No input validation or sanitization is performed. The lack of available tools makes this vulnerability surface effectively harmless.
[FRONTMATTER]: The YAML metadata is used strictly for display and organization. No attempts were found to use frontmatter fields for prompt injection or system configuration bypass.

shorts-strategy