prism-stack

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

• Third-party content exposure detected (high risk: 0.70). The workflow ingests outsider-authored free text when it runs python3 -m prism_agent.uiux search --domain stacks --query "{stack_name}", because the retrieved “stacks” knowledge base content is not shown as operating-user-authored and is likely fetched/returned as readable text into the agent’s LLM context.