Break complex AI requests into smaller sequential prompts.

Use the request dilution technique: split one large request into several small focused prompts and combine the outputs at the end.