paper-reading
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches content from arXiv URLs. This is a well-known academic repository and the behavior is consistent with the skill's purpose.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process untrusted external content from papers. Ingestion points: Local PDF file paths, arXiv URLs, and pasted text specified in SKILL.md. Boundary markers: Absent. The instructions do not provide explicit delimiters or warnings for the agent to ignore instructions embedded within the source documents. Capability inventory: The skill facilitates file system reading (for local PDFs) and network access (for arXiv content) as part of its document processing flow. Sanitization: Absent. There is no evidence of filtering or validation of the content before it is processed for summarization.
Audit Metadata