automation-pilot-catalog-explorer

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md describes runtime API calls (e.g., curl ... /api/v1/catalogs and /api/v1/commands... and /api/v1/commands/$COMMAND_ID) whose JSON fields like description and configuration.executors would be ingested as readable text by the agent, and those catalog/command definitions can be outsider-authored (tenant custom catalogs/commands or SAP-provided content).