workitem

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

• Third-party content exposure detected (high risk: 0.95). Yes—Step 1/2 fetches outsider-authored work item text (GitHub issue bodies/comments via gh issue view ... --comments and ADO work item details via az boards work-item show ...), and Step 2 further reads outsider PR bodies via gh pr view ... --json title,body,..., which the agent will ingest into the LLM context for triage/research.