The Agent Skills Directory

[SAFE]: The skill serves its stated purpose of auditing and rewriting prose to remove formulaic AI patterns. No malicious behavior or security risks were detected.
[DATA_EXFILTRATION]: No network operations or external data transfer mechanisms were found in the skill instructions or reference files. The skill operates locally on provided text.
[COMMAND_EXECUTION]: The skill does not invoke shell commands, spawn subprocesses, or attempt to execute arbitrary scripts on the underlying system.
[OBFUSCATION]: The content is provided in clear, readable Markdown. There are no signs of Base64 encoding, zero-width characters, homoglyphs, or other techniques used to hide malicious intent.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data (user-provided text and files). While this represents a surface for indirect prompt injection, the skill's logic is constrained to stylistic transformations and it lacks the necessary capabilities (e.g., network access or file writing) to facilitate a successful exfiltration or execution attack.

humanizer