okx-mcp-services

Warn

Audited by Snyk on Nov 15, 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 NewsLiquid query and scheduled-query paths return outsider-authored news, Twitter posts, and Meme opinions that the agent ingests, and the scheduled-query path can pick up newly posted matching content without the agent selecting a specific item, so this is at least medium under recall-first scoring.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

  • Direct money access detected (high risk: 0.70). The skill documents wallet-backed plan purchases and renewal for NewsLiquid/6551 (lines 15, 57). Those are operational payments the agent performs through the user's wallet, not just service-call costs. They are direct financial operations (purchase, renewal, payment). Lines 57 and 63-65 describe the controls. Service login is described as separate from payment authority. Consent is obtained through the provider's business Flow, which the agent follows, and wallet isolation is also agent-followed. Neither is a documented authentication or authorization mechanism enforced by the service or execution layer for the payment route. The skill refers to "existing wallet/payment mechanisms" without describing how they authenticate or authorize. Under the evidence standard, instruction-based consent and unspecified wallet mechanisms do not qualify as independent control, so the direct operation has no established qualifying control.

Issues (2)

W011
MEDIUM

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

W009
MEDIUM

Direct money access capability detected (payment gateways, crypto, banking).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Nov 15, 2026, 02:08 PM
Issues
2