The Agent Skills Directory

[SAFE]: The skill is a stylistic editing tool that identifies and removes AI writing patterns based on documented linguistic indicators from Wikipedia. It performs no network operations or sensitive data access. \n- [COMMAND_EXECUTION]: The skill requests standard file manipulation tools (Read, Write, Edit, Grep, Glob). These are appropriate for its function as a writing editor and do not include high-risk commands or privilege escalation attempts. \n- [PROMPT_INJECTION]: Analysis of the instructions shows no attempts to bypass safety filters or override agent behavior. The inclusion of self-critique prompts is a functional part of the text-improvement workflow. \n- [DATA_EXFILTRATION]: No network tools or patterns indicative of data exfiltration were found. The skill does not access sensitive directories or hardcoded credentials.

humanizer