agora

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's "Proxy requests" section instructs using Membrane to send arbitrary requests to the Agora API (membrane request CONNECTION_ID /path/to/endpoint) and the skill also references transcripts/participants, so the agent can fetch and directly ingest user-generated third-party content (meeting transcripts/messages) that could influence subsequent actions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 4, 2026, 08:55 AM