admin

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

• Third-party content exposure detected (high risk: 0.80). The skill's docs explicitly require agents to call external MCP servers (e.g., SimpleMem at mem.self-host.io via memory_query) and list web‑facing MCP servers like "fetch", "brave-search", and "puppeteer" in the MCP references (references/memory-integration.md and references/mcp.md), and agents are instructed to use those external memory/search/fetch tools before operations — meaning untrusted third‑party content can be ingested and can influence decisions and next actions.