3scribe

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md shows the agent uses Membrane to connect to 3Scribe and run actions such as "fetch-transcript" (see "Popular actions" and "Running actions"), which clearly retrieves user-generated third-party transcript content that the agent is expected to read and could influence subsequent actions.