assembla

Warn

Audited by Snyk on Mar 4, 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). The SKILL.md explicitly exposes fetching Assembla user-generated content (e.g., "List Ticket Comments", "Get Ticket", "Wiki Page") and allows proxying API requests via the Membrane CLI, so untrusted third‑party text from Assembla spaces could be ingested and influence the agent's decisions or subsequent tool use.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 4, 2026, 08:59 AM