arxiv-package

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a legitimate developer utility for automating the preparation of academic manuscripts for submission. It does not contain any malicious patterns or hidden code.
  • [COMMAND_EXECUTION]: The skill utilizes standard system commands such as tar and zip to bundle local project files. The instructions include a validation step to ensure no filenames contain spaces or special characters, which mitigates potential command injection risks.
  • [DATA_EXFILTRATION]: The skill promotes positive security practices by explicitly instructing the agent to scan for and exclude sensitive files such as .env, credentials, private notes, and version control directories (.git).
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 02:40 PM