exploratory-data-analysis

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly ingests arbitrary user-provided scientific data files (user-supplied file paths processed by scripts/eda_analyzer.py and the Step 3 analysis workflow) and reads their contents to generate reports, which exposes the agent to untrusted third-party content that could carry indirect prompt-injection payloads.