orient-gaps

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

• Third-party content exposure detected (medium risk: 0.65). SKILL.md/protocol.yon require a bounded subagent to gather evidence from the provided scope (git tier, plan/doc tier, conversation/folder tier), which at runtime can ingest outsider-authored free text such as issue/PR/wiki/discussion body text or other non-user content into the LLM context via the subagent’s tool calls.