tubular

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The SKILL.md instructs using Membrane to run actions and a proxy request (e.g., "membrane request CONNECTION_ID /path/to/endpoint" and action runs) against the external Tubular API, which can return user-generated social media content (comments) from third-party sources that the agent is expected to read and act on.