stash-ai-memory

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests arbitrary user-generated episodes via the /api/episodes endpoint and exposes them through /api/recall and the "Injecting Memory into System Prompt" pattern, so untrusted third‑party content stored as memories can be read and directly influence agent prompts and actions.