peer-review-writer
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Analysis of the skill instructions and supporting reference documents found no evidence of malicious patterns, obfuscation, or security vulnerabilities.
- [DATA_EXFILTRATION]: The skill includes explicit security guidelines regarding confidentiality, instructing the agent never to leak manuscripts or attempt to deanonymize authors.
- [PROMPT_INJECTION]: The instructions contain clear usage boundaries (e.g., separating reviewer and author roles) without any patterns attempting to bypass agent safety filters or system instructions.
- [EXTERNAL_DOWNLOADS]: All external references are limited to academic citations (arXiv IDs) used for grounding. No automated network downloads or script execution patterns were identified.
- [COMMAND_EXECUTION]: The skill does not utilize shell commands or local file system operations beyond reading its own reference files.
Audit Metadata