citation-diversifier
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input files such as DRAFT.md and outline.yml. These external data sources create a surface for indirect prompt injection where malicious instructions embedded in the document text could theoretically influence the AI agent's behavior during citation analysis and report generation.
Audit Metadata