single-cell-annotation

Warn

Audited by Snyk on May 2, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly instructs downloading and using public third-party models and reference datasets (e.g., "Download immune cell model" via celltypist models.Model.load in Step 3 and selecting tissue-matched references such as the Human Cell Atlas/Tabula Muris in Step 4), so the agent ingests open, untrusted external content that can materially influence annotation decisions.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
May 2, 2026, 02:05 PM
Issues
1