The Prompt-to-Paper framework uses a multi-agent system to generate scientific manuscripts grounded in verifiable literature. It replaces fabricated results with deterministic retrieval-augmented generation and section-aware relevance scoring. This architecture targets the rigor gap in AI-written papers. Researchers can now automate drafting while maintaining strict grounding in existing bioinformatics citations.