evolink-nano-banana-2
Warn
Audited by Snyk on Feb 28, 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 explicitly accepts and fetches arbitrary external image URLs / file_url (see SKILL.md Step 2 & Step 4 "image_urls" / references/api-params.md upload_file and image_urls parameters, and scripts/evolink-image-gen.sh --image) and uses those untrusted, user-provided third-party files as reference input for generation, which can materially influence model behavior and tool actions.
Audit Metadata