This study presents optimized sgRNA design rules to enhance the efficiency and accuracy of CRISPR-Cas9-based genetic screens. The authors developed human and mouse genome-wide libraries using previously established rules for predicting on-target efficiency. They performed positive and negative selection screens and observed improved results with these optimized libraries. They also profiled the off-target activity of thousands of sgRNAs and developed a metric to predict off-target sites. By incorporating findings from large-scale empirical data, they improved computational design rules and created optimized sgRNA libraries that maximize on-target activity and minimize off-target effects, enabling more effective and efficient genetic screens and genome engineering. The study demonstrates that the Avana and Asiago libraries, designed based on these optimized rules, outperformed existing libraries in identifying genes involved in vemurafenib resistance and other phenotypes. The Avana library identified more PanCancer genes than either version of the GeCKO library. The authors also developed a new off-target scoring metric, the CFD score, which outperformed existing metrics in predicting off-target activity. The CFD score was validated using multiple datasets and showed high correlation with experimental data. The study also highlights the importance of specificity in sgRNA design, as off-target effects can significantly impact screening results. The authors developed Rule Set 2, which improves on-target activity predictions and outperforms Rule Set 1 in multiple datasets. The optimized libraries, Brunello and Brie, were designed to maximize Rule Set 2 scores and minimize off-target sites, representing a clear improvement over existing libraries. The study provides a resource for the design of improved sgRNA reagents for large-scale screens and small-scale gene editing experiments. The authors emphasize the importance of using multiple cellular models to confirm the generalizability of findings and suggest that future work will determine if results obtained using Streptococcus pyogenes Cas9 provide useful lessons regarding the activity of other Cas9 proteins. The experimental and analytical approaches described here illustrate a powerful method to uncover factors contributing to sgRNA activity and specificity and to optimize reagent design for large-scale functional genomics.This study presents optimized sgRNA design rules to enhance the efficiency and accuracy of CRISPR-Cas9-based genetic screens. The authors developed human and mouse genome-wide libraries using previously established rules for predicting on-target efficiency. They performed positive and negative selection screens and observed improved results with these optimized libraries. They also profiled the off-target activity of thousands of sgRNAs and developed a metric to predict off-target sites. By incorporating findings from large-scale empirical data, they improved computational design rules and created optimized sgRNA libraries that maximize on-target activity and minimize off-target effects, enabling more effective and efficient genetic screens and genome engineering. The study demonstrates that the Avana and Asiago libraries, designed based on these optimized rules, outperformed existing libraries in identifying genes involved in vemurafenib resistance and other phenotypes. The Avana library identified more PanCancer genes than either version of the GeCKO library. The authors also developed a new off-target scoring metric, the CFD score, which outperformed existing metrics in predicting off-target activity. The CFD score was validated using multiple datasets and showed high correlation with experimental data. The study also highlights the importance of specificity in sgRNA design, as off-target effects can significantly impact screening results. The authors developed Rule Set 2, which improves on-target activity predictions and outperforms Rule Set 1 in multiple datasets. The optimized libraries, Brunello and Brie, were designed to maximize Rule Set 2 scores and minimize off-target sites, representing a clear improvement over existing libraries. The study provides a resource for the design of improved sgRNA reagents for large-scale screens and small-scale gene editing experiments. The authors emphasize the importance of using multiple cellular models to confirm the generalizability of findings and suggest that future work will determine if results obtained using Streptococcus pyogenes Cas9 provide useful lessons regarding the activity of other Cas9 proteins. The experimental and analytical approaches described here illustrate a powerful method to uncover factors contributing to sgRNA activity and specificity and to optimize reagent design for large-scale functional genomics.