cross-pollinator
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches scientific data from the official arXiv API (
export.arxiv.org). This is a well-known service and the usage is consistent with the skill's research-oriented purpose. - [PROMPT_INJECTION]: The skill demonstrates an attack surface for indirect prompt injection as it processes untrusted external content from paper abstracts. However, the risk is negligible due to the following factors:
- Ingestion points: Data enters the context via the arXiv API in Step 2 of the methodology.
- Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the fetched abstracts.
- Capability inventory: The skill does not have any dangerous capabilities, such as file system writes, subprocess execution, or network exfiltration to arbitrary domains.
- Sanitization: No explicit sanitization or filtering of the external data is described.
Audit Metadata