code-review-recent-changes

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

• Third-party content exposure detected (high risk: 0.70). The required workflow can fetch and read outsider-authored issue/PRD content at runtime (e.g., via gh issue view <number> --comments or reading docs/issues//backlog entries that may be written by others), and then passes that fetched free text into the Spec sub-agent’s LLM context.