bkd

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests user-generated follow-up prompts and issue logs from a BKD server (e.g., POST/GET to $BKD_URL/projects/{projectId}/issues/{issueId}/follow-up and GET $BKD_URL/projects/{projectId}/issues/{issueId}/logs/filter/... as described in SKILL.md and references/orchestration.md and references/quality-review.md), and those third-party follow-up/log contents are parsed and used to drive decisions and actions (restarting work, merging, status transitions), so untrusted content can materially influence agent behavior.