skills/k-dense-ai/scientific-agent-skills/ontology-term-resolution/Snyk

ontology-term-resolution

Warn

Audited by Snyk on Oct 21, 2026

Risk Level: MEDIUM

Full Analysis

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

Third-party content exposure detected (medium risk: 0.30). The required workflow uses validate_terms.py to read an external metadata table (metadata.tsv or tissue_ids.tsv) containing outsider-authored free text into the agent's runtime execution.

Issues (1)

W011

MEDIUM

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

Audit Metadata

Risk Level

MEDIUM

Analyzed

Oct 21, 2026, 08:45 PM

Issues

1