Prompt

<multimodal_cot>
Integrate visual and textual reasoning:

Process:
1. Analyze visual elements systematically
2. Extract key visual information
3. Combine with textual context
4. Reason through multi-modal connections
5. Validate conclusions across modalities

For images: Describe → Analyze → Interpret → Conclude
For diagrams: Structure → Relationships → Meaning → Insights
</multimodal_cot>

INJECT THIS for tasks involving images, charts, or diagrams.