Update SKILL.md files to add double quotation marks for all skills, ensuring clarity and consistency across all entries.

scientific-packages/statsmodels/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: statsmodels
-description: Comprehensive statistical modeling and econometric analysis toolkit for Python. This skill should be used when you need to fit statistical models, perform hypothesis testing, conduct econometric analysis, or analyze time series data. Specifically, use this skill for: (1) Linear regression analysis including OLS, WLS, GLS, quantile regression, and mixed effects models for continuous outcomes with comprehensive diagnostics, robust standard errors, and influence analysis; (2) Generalized linear models (GLM) for non-normal outcomes including logistic regression for binary data, Poisson/Negative Binomial for count data, and Gamma regression for skewed continuous data; (3) Discrete choice models including logit/probit for binary outcomes, multinomial logit for categorical outcomes, and zero-inflated models for count data with excess zeros; (4) Time series analysis including ARIMA/SARIMAX for univariate forecasting, VAR/VARMAX for multivariate analysis, state space models, and exponential smoothing with stationarity testing and residual diagnostics; (5) Statistical testing and diagnostics including tests for heteroskedasticity, autocorrelation, normality, multicollinearity, influence detection, ANOVA, and hypothesis testing with appropriate corrections; (6) Model comparison and selection using AIC/BIC, likelihood ratio tests, and cross-validation; (7) Causal inference with instrumental variables, difference-in-differences, and regression discontinuity designs. This skill is essential when you need rigorous statistical inference, publication-ready results, detailed model diagnostics, or when working with econometric, biomedical, or social science data requiring proper statistical methodology. Always use this skill instead of basic regression when you need confidence intervals, p-values, diagnostic tests, or when assumptions need to be validated.
+description: "Comprehensive statistical modeling and econometric analysis toolkit for Python. This skill should be used when you need to fit statistical models, perform hypothesis testing, conduct econometric analysis, or analyze time series data. Specifically, use this skill for: (1) Linear regression analysis including OLS, WLS, GLS, quantile regression, and mixed effects models for continuous outcomes with comprehensive diagnostics, robust standard errors, and influence analysis; (2) Generalized linear models (GLM) for non-normal outcomes including logistic regression for binary data, Poisson/Negative Binomial for count data, and Gamma regression for skewed continuous data; (3) Discrete choice models including logit/probit for binary outcomes, multinomial logit for categorical outcomes, and zero-inflated models for count data with excess zeros; (4) Time series analysis including ARIMA/SARIMAX for univariate forecasting, VAR/VARMAX for multivariate analysis, state space models, and exponential smoothing with stationarity testing and residual diagnostics; (5) Statistical testing and diagnostics including tests for heteroskedasticity, autocorrelation, normality, multicollinearity, influence detection, ANOVA, and hypothesis testing with appropriate corrections; (6) Model comparison and selection using AIC/BIC, likelihood ratio tests, and cross-validation; (7) Causal inference with instrumental variables, difference-in-differences, and regression discontinuity designs. This skill is essential when you need rigorous statistical inference, publication-ready results, detailed model diagnostics, or when working with econometric, biomedical, or social science data requiring proper statistical methodology. Always use this skill instead of basic regression when you need confidence intervals, p-values, diagnostic tests, or when assumptions need to be validated."
 ---
 
 # Statsmodels: Statistical Modeling and Econometrics