character-passport

Warn

Audited by Snyk on Nov 16, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (low risk: 0.10). The required workflow ingests outsider-authored text only when the agent actively searches for or selects real-world reference images, costumes and accents (reference-sites.md and the find-references skill), which is a low-risk pull path; the core passport, sheet, voice and prompt steps use the user's own moodboard and script, varg generation, and structured data from Poly Haven and physicallybased.info.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (medium risk: 0.30). The page fetches the physicallybased.info materials API at runtime and uses its returned values to set the prop's brass material colour and metalness, so the content configures rendering; that publisher is an individual-owned project (unofficial, reputable), which makes it medium, while the three.js, three-vrm, Poly Haven, Tencent and Microsoft sources are official and do not raise the tier.

Issues (2)

W011
MEDIUM

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

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 16, 2026, 08:47 PM
Issues
2