review-paper
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's behavior is entirely consistent with its described purpose. No hardcoded credentials, malicious network requests, or unauthorized system modifications were found.\n- [PROMPT_INJECTION]: Evaluated the risk of indirect prompt injection from manuscript content. The risk is assessed as safe because the skill does not have access to dangerous capabilities such as command execution or internet access, and processing manuscript data is core to the skill's primary function.\n
- Ingestion points: Manuscript files provided via
$ARGUMENTSor located inmaster_supporting_docs/.\n - Boundary markers: The prompt instructions do not utilize delimiters or specific instructions to ignore embedded commands within the ingested paper content.\n
- Capability inventory: Tools are limited to
Read,Write,Grep,Glob, andTask. No shell execution or network capabilities are present.\n - Sanitization: The skill reads the raw content of papers without filtering for adversarial instructions.
Audit Metadata