openspec-linearized

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

• Third-party content exposure detected (high risk: 0.75). The required workflow explicitly reads outsider-authored Linear issue/card content (title/description/labels/comments/links) via Linear MCP tools like list_issues/list_comments during proposal discovery and uses that text as the “business brief,” which becomes LLM-readable context; this is an indirect prompt-injection risk.