Security audit
Security checks for vulnerabilities and agentic risk
This Chinese novel-writing skill is mostly purpose-aligned, but it includes unattended hourly writing automation and local file changes that users should review before installing.
Install only if you want this skill to manage a Chinese novel project in the disclosed paths. Before using any scheduled mode, confirm where it will write, how to stop it, and whether each generated chapter and outline update will be reviewed before being saved.
The skill advertises very broad activation phrases for generic writing tasks, which can cause it to trigger in situations where the user only wants ordinary writing help. Because the skill also contains automatic file read/write behavior, overbroad activation increases the chance of unintended local file access or modification without clear user intent.
These instructions direct automatic reads from and writes to multiple local paths, including home-directory and Desktop files, without any warning, consent gate, or confirmation step. This creates a real risk of unintended data exposure, overwriting user content, and exfiltration of sensitive material from unrelated local files under the guise of writing assistance.
The hourly auto-execution instruction enables repeated unattended file modifications, which can amplify accidental corruption, overwrite risks, and unexpected resource use over time. Scheduled autonomous writes are especially dangerous in a skill because they can continue changing local content after the initial interaction without fresh user review.
The skill description and title are written entirely in Chinese and present the skill as operating in that language, but they do not state that Chinese is optional or limited to a justified region-specific use case. This can conflict with language-choice policy when users are not given an explicit opt-in or alternative.
No suspicious patterns detected.
