Security audit
Security checks for vulnerabilities and agentic risk
This skill is a straightforward text-cleanup helper that can modify user-selected files, with no evidence of hidden network access, persistence, credential use, or unrelated behavior.
Install this if you want an agent to clean up AI-like writing patterns. Use care on important documents because the skill may edit files in place when asked to process files; ask for a diff or approval step before overwriting valuable content.
The skill explicitly instructs the agent to read a file, transform it, and write it back, but it declares no tool scope or allowed-tools restrictions. That mismatch can let a broadly invoked editing skill perform file modifications without clear capability boundaries, increasing the chance of unintended or over-broad file writes.
The trigger phrases are very broad and overlap with normal editing requests like 'make this sound human' or 'clean up AI writing.' This can cause the skill to activate in contexts the user did not intend, which is more risky here because the skill also describes file-modifying behavior.
The skill says that for files it should read the file, apply transformations, and write it back, but it does not present a prominent warning that user files may be modified. Without an explicit safety check, users may not realize the skill can overwrite content, especially since its primary framing is simple text cleanup.
No suspicious patterns detected.
