aliyun-arms-query

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 ARMS API responses (e.g., get_trace_detail.py/search_traces.py call client.get_trace() and then render/print tag_entry_list and log_event_list values, which are untrusted trace/span log text authored by external systems/users), and those printed strings are typically ingested into the agent’s LLM context as evidence.