autoverse-topic-prep
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured instructions for academic research and paper analysis using specific tools and authoritative sources.
- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates an inherent attack surface for indirect prompt injection due to its core function of processing external web and academic content.
- Ingestion points: Processes data from the
autoverseCLI tool and external web search results including DOI pages, publisher websites, PubMed, and arXiv (SKILL.md). - Boundary markers: The instructions do not specify explicit delimiters or 'ignore instructions' directives for handling the retrieved external content.
- Capability inventory: The skill utilizes the
autoversetool for search and resolution, web searching for verification, and file system operations for generating research outputs like Word, PDF, and BibTeX files. - Sanitization: No specific sanitization or validation logic is defined for the external text before it is processed by the agent.
- Note: This surface is tied to the skill's primary research purpose and targets well-known academic repositories.
Audit Metadata