epic-executor

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

• Third-party content exposure detected (medium risk: 0.65). Yes—during Step 1 the skill calls tracker adapter operations like discoverChildren(epic-id) and getDependencies(id) (e.g., JIRA/Linear/ADO/local), which can ingest outsider-authored free text from issue/ticket fields (summaries/descriptions/comments) into the agent context for planning/execution and then into task-executor.