The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits susceptibility to indirect prompt injection (Category 8) by processing untrusted data with file-modification capabilities.
Ingestion points: The skill is designed to ingest and process arbitrary text provided by users for the purpose of 'humanizing' it.
Boundary markers: The instructions lack explicit delimiters or specific directives for the agent to ignore any embedded instructions or commands within the provided text.
Capability inventory: The skill utilizes Write and Edit tools, allowing it to modify files based on the output of the processing task.
Sanitization: No input validation or sanitization of the processed text is mentioned in the instructions.
[SAFE]: The skill's primary focus is on linguistic patterns and stylistic rewriting, based on legitimate community-maintained resources such as WikiProject AI Cleanup.
[SAFE]: No patterns associated with data exfiltration, remote code execution, or obfuscation were found in the skill content.

humanizer-zh