loyaltylion

Warn

Audited by Snyk on Apr 23, 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). This skill explicitly fetches and ingests data from a third-party service (via Membrane actions and the "membrane request CONNECTION_ID /path/to/endpoint" proxy to the LoyaltyLion API), so the agent will read external, potentially user-generated/untrusted content from that API which could influence subsequent tool calls and decisions.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 23, 2026, 12:54 AM
Issues
1