primekg

Warn

Audited by Snyk on Jun 14, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.75). Outsider content is ingested indirectly from the PrimeKG CSV dataset at runtime (_load_kg() reads kg.csv from a local path), and that dataset is not authored by the operating user (it aggregates external databases/literature), so its text fields (e.g., node/neighbor names and sources) can be fed into the agent’s LLM context via search_nodes() / get_neighbors() / get_disease_context() outputs.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 14, 2026, 09:22 AM
Issues
1
Security Audit — snyk — primekg