langsmith

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs fetching shared, user-contributed prompts via the public Prompt Hub (see "Step 5: Prompt Hub and Annotation Queues" — hub.pull("rlm/rag-prompt")), which ingests untrusted third-party prompt content that the agent would read and could materially change its behavior.