compile-generation-prompts

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

• Third-party content exposure detected (medium risk: 0.30). The skill runtime compiles the incoming payload (validated against schemas/input.schema.json) into provider_requests prompts/parameters, so any outsider-authored text inside payload can directly reach the model-free-text compilation output path without requiring fetching or searching a specific externally selected item.