clean
Warn
Audited by Snyk on Jul 27, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.65). The required workflow reads user-supplied dataset files from
input/<dataset>/at runtime (e.g., arbitrary CSV/XLSX/JSON content), and that file body text is parsed into a pandas DataFrame and then used to generate LLM-visible artifacts like the cleaning summary/log—so if those files contain outsider-provided free text, it can be ingested into the agent context indirectly.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata