alphaxiv

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches paper metadata and structured overviews from alphaxiv.org and arxiv.org. These external calls are expected as they are central to the skill's purpose of research assistance.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests untrusted content from the web and processes it without explicit isolation. 1. Ingestion points: Untrusted data is retrieved from alphaxiv.org/api and arxiv.org/abs (SKILL.md). 2. Boundary markers: The prompt template does not use delimiters or instructions to ignore potential commands embedded within the paper summaries. 3. Capability inventory: The skill has access to WebFetch (network) and Write (filesystem) capabilities. 4. Sanitization: There is no mention of sanitizing or validating the fetched external content before it is displayed or saved to the user's research notes.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 05:59 AM
Security Audit — agent-trust-hub — alphaxiv