LM Studio

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly documents configuring and using remote MCP servers (see mcp-playbooks.md and the mcp.json "mcpServers" pattern), which are third-party/untrusted endpoints whose tool outputs the agent is expected to run/read and that can materially influence subsequent tool use and decisions, creating an indirect prompt-injection risk.