experiment-analzyer

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs the agent to fetch and include raw span/event inputs, outputs, and "actual prompt/code snippets" (via tools like get_llmobs_span_content and event details), which would require reproducing any secrets contained in those spans verbatim, creating an exfiltration risk.