kernel-tileir-optimization

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

• Third-party content exposure detected (medium risk: 0.30). In scripts/classify_kernel.py, outsider-provided kernel source text is ingested directly via the --code flag or by reading a user-supplied --file and then analyzed/modified by classify_kernel() and apply_optimizations(), meaning the workflow can process arbitrary free text provided by a user without selecting a trusted item first.