server-log-crawler-analyst

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

• Third-party content exposure detected (medium risk: 0.30). In the “Identify available evidence” step, the workflow ingests provided access-log snippets or exported summaries (including outsider-controlled text pasted/uploaded by the user) and runs LLM-based summarization/classification over them, which could be indirectly influenced by prompt-injection strings embedded in that log content.