The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use standard system utilities including zip, unzip, and ln for packaging and verifying skill components. These operations are conducted within the user's local ~/.claude/skills/ directory and are consistent with the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill features a capability to ingest external text (STYLE_PROFILE.md) and incorporate it into newly generated instructions.
Ingestion points: Reads STYLE_PROFILE.md from the local workspace or a user-provided path.
Boundary markers: The generated SKILL.md template lacks explicit delimiters or warnings to isolate processed content from instructions.
Capability inventory: The skill possesses file system write access and the ability to execute compression and linking commands.
Sanitization: No sanitization or escaping of the input profile content is performed during the template generation process.

style-to-skill