The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requests access to powerful file system tools including Read, Write, Edit, Glob, and Grep. These capabilities are used for the intended purpose of scanning and editing text files within the user's workspace to improve writing style.
[PROMPT_INJECTION]: As a text-processing utility, the skill has a surface for indirect prompt injection (Category 8). It ingests untrusted user text and processes it using instructions that could be manipulated. However, the skill instructions are focused on stylistic editing, and the potential for abuse is mitigated by the agent's underlying safety guardrails and the legitimate nature of the utility.
Ingestion points: Processes text provided via the /humanizer command or within the chat context.
Boundary markers: No specific delimiters or boundary markers are defined in the instructions to separate user data from instructions.
Capability inventory: Access to Read, Write, Edit, Glob, and Grep tools.
Sanitization: No explicit sanitization or validation of the input text is performed before processing.

humanizer