review

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

• Third-party content exposure detected (medium risk: 0.30). On /pi:review, the skill ingests outsider-authored free text from the user’s own command arguments as TASK_TEXT (free-text task description) and also passes git diff / git context / CLAUDE.md into pi as system-prompt context, meaning the LLM runtime will read user-provided text.