client-memory-synthesizer

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text can enter the LLM context via the skill’s runtime --input client.json / memory_inputs fields (e.g., SOWs, QBR outputs, meeting notes, email threads, escalations) that may include outsider-authored content; the script then LLM-generates result.tenant_context from those strings and writes/pastes it into Claude Project Instructions.