rocks - in R, just

  library(quantmod)
  getSymbols("CPIAUCNS", src="FRED")

then you can get nice graphs with something like

  library(ggplot2)
  CPIPCT = data.frame(CPIAUCNS / lag(CPIAUCNS,12) * 100 -100)
  CPIPCT$date= as.Date(rownames(CPIPCT))
  CPIPCT <- tail(CPIPCT, 240)
  CPI.m <- melt(CPIPCT, id = 'date')
  ggplot(data=CPI.m, aes(x=date, y=value, colour=variable)) + ylab("YoY % Change") + theme_bw() + scale_x_date("")  + geom_line(size=1.4) + opts(plot.background = theme_rect(colour = 'black', fill = '#CCCCEE', size = 1, linetype='solid'), legend.position="top") + scale_colour_manual("CPI", breaks=c("CPIAUCNS"), values = c("#0000CC"), labels=c('All Urban'))

or just get the graphs online with Fred Graph -

http://research.stlouisfed.org/fred2/graph/