discussions

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow calls the Protocols.io Discussions API endpoints (e.g., GET /protocols/{protocol_id}/comments and GET /protocols/{protocol_id}/steps/{step_id}/comments) which return comment content authored by other users, and that free text is then ingested into the agent’s LLM context for analysis/response.