internal-critique
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of academic papers, thesis chapters, or drafts.
- Ingestion points: The skill processes text provided by the user (papers/drafts) as the primary input for its review protocol.
- Boundary markers: The instructions do not define clear delimiters or use specialized instructions to help the agent distinguish between the paper's content and the review instructions.
- Capability inventory: The skill's capabilities are limited to generating a structured markdown report. It does not request access to any system tools, network operations, or file-writing capabilities.
- Sanitization: There is no mechanism described for sanitizing or escaping potentially malicious instructions embedded within the academic drafts.
Audit Metadata