multi-version-compliance

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime LLM context is populated from Linear task content via the Linear MCP calls (mcp__linear-server__get_issue and mcp__linear-server__list_comments), and Linear issues/comments are outsider-authored free text (not authored by the operating user), which can include prompt-injection payloads.