arxiv-translator

Warn

Audited by Snyk on Aug 8, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (low risk: 0.10). The skill downloads arXiv-provided LaTeX source/PDF for a specific arXiv paper and then ingests the resulting .tex and compilation .log files locally (via curl/tar + split_tex.py + reading .log), but this text is first-party arXiv content selected from an explicit paper ID rather than outsider-authored free text submitted into a monitored queue/feed the workflow reads without targeting.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.70). The skill instructs downloading arXiv source at runtime (e.g., https://arxiv.org/src/xxxx.xxxxx) and then running local compilation (xelatex/bibtex) on that fetched LaTeX, which results in executing remote content via the TeX engine.

Issues (2)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 8, 2026, 03:02 AM
Issues
2
Security Audit — snyk — arxiv-translator