The Agent Skills Directory

[SAFE]: No security issues detected. The skill primarily consists of instructional guidelines and local reference documents for simulating a specific academic review style. The Chinese text included in the skill description and triggers represents legitimate academic terminology for the tool's intended use-case (peer review simulation).
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user data, specifically manuscript drafts and abstracts. While this provides a surface for indirect prompt injection, the skill possesses no dangerous capabilities: it does not perform network operations, file system writes, or shell command executions. Consequently, even if a user provided a malicious manuscript, there is no mechanism for the skill to perform harmful actions.

nature-reviewer