sourcing-and-fulfillment

Fail

Audited by Snyk on Nov 16, 2026

Risk Level: CRITICAL
Full Analysis

CRITICAL E004: Prompt injection detected in skill instructions.

  • Potential prompt injection detected (medium risk: 0.30). The skill's only self-serving element is a disclosed Whop referral code, applied to onboarding and new-business creation, which is a plausible but openly stated monetization that falls under the medium "unexpected action / promotion" category rather than any sign of malicious intent against the user.

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

  • Third-party content exposure detected (medium risk: 0.30). The required RFQ comparison workflow reads supplier-authored replies to requests for quotation, and supplier discovery reads supplier chat replies to listing questions. Outside suppliers can send these replies without the agent first selecting a specific item, so the text reaches the agent as outsider-authored content.

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

  • Potentially malicious external URL detected (medium risk: 0.30). The only inbound operational source is the raw GitHub README that SKILL.md directs the agent to use for other skills' installation commands, which is a mutable main-branch file from the project's own organization and not clearly recognized first-party provenance, so it is medium; the Whop installers, Whop/CJ/Shopify/DSers docs, and the Vercel reference are strong or official or not executed.

Issues (3)

E004
CRITICAL

Prompt injection detected in skill instructions.

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
CRITICAL
Analyzed
Nov 16, 2026, 10:47 AM
Issues
3