paper-review

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's logic is confined to document analysis and academic critiquing. It does not attempt to access credentials, modify the file system, or establish persistence.
  • [PROMPT_INJECTION]: Although the skill processes external papers (untrusted data), it lacks high-privilege capabilities that would make indirect prompt injection a significant risk.
  • Ingestion points: Manuscript drafts and final submissions are processed in the review workflow.
  • Boundary markers: Instructions do not specify delimiters to isolate the manuscript text.
  • Capability inventory: The skill utilizes a WebSearch tool for literature review.
  • Sanitization: No sanitization logic for the ingested manuscript content is included in the skill definition.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 01:22 AM