Researchers introduce CAFES framework to address context bottleneck
A new paper from DX, Capital One, GitHub, University of Victoria, and Google argues that context quality is the central bottleneck in agentic workflows. The CAFES framework introduces five properties to help teams build more effective agents.
Researchers from DX, Capital One, GitHub, the University of Victoria, and Google have co-authored a paper introducing the CAFES framework. The work explores why context quality has become the central bottleneck in agentic workflows. The core argument is that when teams provide better context, they build more effective agents and achieve significantly better engineering outcomes. The paper introduces five properties designed to help teams structure their context to overcome this constraint. The researchers argue that context quality is the primary limit on agentic workflow performance and that the CAFES framework provides a necessary structure to address this issue. The paper is available for reading via getdx.com/cafes.