range
Warn
Audited by Snyk on Mar 13, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). This skill instructs the agent to fetch and act on Range data via the Membrane CLI (e.g., "membrane action run ..." and "membrane request CONNECTION_ID /path/to/endpoint"), which ingests third-party, user-generated Range content that the agent would read and could influence subsequent actions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata