The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted user-provided text to perform its humanization tasks.\n- Ingestion points: User input or project files read into the agent context (SKILL.md).\n- Boundary markers: Absent; the instructions do not define clear delimiters or include warnings to ignore instructions embedded within the user text.\n- Capability inventory: The skill leverages Write, Edit, Grep, and Glob tools (SKILL.md), which provide file system modification capabilities.\n- Sanitization: Absent; there is no evidence of input validation or sanitization for the text provided for processing.\n- [SAFE]: The skill references and implements writing guidelines from Wikipedia's 'Signs of AI writing' guide, using a well-known and reputable source for its editing logic.

humanizer