remove-ai-marks
Warn
Audited by Snyk on Nov 11, 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 reads user-supplied local files, but the optional
wm-audit-sitepath (SKILL.md Classify table) actively fetches arbitrary sitemap URLs and inspects their outsider-authored content, so it is low exposure because the agent must choose to run it.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). The reverse-SynthID repository is an individual-owned GitHub project that the setup script (
setup_synthid.sh) and scorer path use at runtime, so it is an operational dependency with unofficial provenance and no concrete warning sign, which is medium; the MarkDiffusion checkout is a pinned, first-party academic project and is not flagged.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata