The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted text provided by the user and has the capability to write and edit files, which creates a potential surface for indirect prompt injection.\n
Ingestion points: The skill ingests arbitrary text for 'humanization' as defined in the task instructions in SKILL.md.\n
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the input text.\n
Capability inventory: The skill frontmatter defines access to Read, Write, Edit, Grep, and Glob tools, enabling file modifications.\n
Sanitization: No sanitization or validation logic is specified for the input text.

humanizer