epic-runner

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

• Third-party content exposure detected (medium risk: 0.30). In Phase 0/1 the skill reads externally authored issue text from the user-specified epic reference via tracker bindings (e.g., GitHub gh issue list/gh issue view pulls each issue body/task text), so outsider free text can enter the LLM context before it selects specific items.