The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown intended to guide the behavior of an AI agent for text editing tasks. It does not request or utilize any external tools, APIs, or script execution environments.- [REMOTE_CODE_EXECUTION]: There are no mechanisms for downloading or running remote code. The reference to the 'brandonwise/humanizer' GitHub repository is provided as an attribution for research and does not involve automated fetching or execution of external assets.- [DATA_EXFILTRATION]: No network activity or data exfiltration patterns were detected. The skill operates solely on the text provided within the agent session.- [COMMAND_EXECUTION]: The skill does not contain any shell commands, subprocess invocations, or requests to modify the system environment.- [PROMPT_INJECTION]: Although the skill is designed to ingest and transform untrusted user input (the text to be edited), it lacks the necessary capabilities (such as file writes or network access) to be leveraged for harmful indirect prompt injection attacks.

avoid-ai-writing