Writing Papers
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves academic paper content and metadata from external sources including ArXiv, Anna's Archive, and web search tools like Exa Search. ArXiv is recognized as a well-known scholarly service.
- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because it processes untrusted data from external documents and search results. * Ingestion points: Data and text retrieved from ArXiv, Anna's Archive, and web searches as described in the Sources section of SKILL.md. * Boundary markers: Not used; there are no instructions to use delimiters or warnings to ignore instructions within the fetched data. * Capability inventory: The skill has the capability to fetch data from the network and generate complex HTML artifacts. * Sanitization: None; there is no mention of sanitizing or validating the retrieved text before it is used by subagents to extract information.
Audit Metadata