The Agent Skills Directory

[SAFE]: The skill monitors academic literature from established platforms like arXiv, MathSciNet, and Semantic Scholar. It does not utilize any high-risk capabilities such as arbitrary command execution, privilege escalation, or persistence mechanisms.
[SAFE]: No hardcoded credentials or data exfiltration patterns were detected. All network operations are directed at well-known academic services for literature metadata.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (paper abstracts and metadata). However, the risk is negligible because the skill's output is restricted to Markdown digests and it lacks the permissions to perform actions that could be triggered by malicious content in an abstract, such as modifying the file system or executing shell commands.

math-literature-track