triage-issue

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

• Third-party content exposure detected (high risk: 1.00). Outsider-authored free text from the GitHub issue title/body and comments is fetched at runtime via gh api repos/getsentry/action-release/issues/<number> and /comments, written to issue.json/comments.json, then passed into detect_prompt_injection.py (and later used to generate the triage report) as LLM-readable text.