triage

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

• Third-party content exposure detected (high risk: 0.85). Yes—during Phase 1/2/3 the skill ingests outsider-authored issue/card text (e.g., GitHub issue title/body or Trello card desc) and passes it into downstream LLM tool calls (brainstorming/writing-plans/scorecard) as part of the args/prompt context.