Researchers introduced Spark-to-Paper, a system that automates end-to-end research paper generation by implementing thirteen composable skills within an existing coding assistant. The system separates model-based judgments from verifiable operations, coordinating literature retrieval, experiment design, and manuscript revision to keep written claims aligned with empirical evidence. Across eight research topics, the approach achieved 99.5% citation validity and 96.4% figure editability, and substantially improved detection of fabricated content compared to a baseline system through integrated integrity checks and review mechanisms.
