Rocks!

libressl
This was created as a response to crazy levels of pointless complexity (like reimplementing their own malloc system) in the core cryptography library almost all linux and bsd programs use openssl. They cleaned all the code up and untangled it.


 * http://www.libressl.org/

signify
A minimal tool to sign documents and verify signed documents, replaces and improves gpg.


 * http://www.openbsd.org/papers/bsdcan-signify.html

musl libc
This is a complete reimplementation of GNU libc with a focus on high quality simple implementation. A few linux distros are using it completely. Having a second alternative implementation of libc helps pull things back to the POSIX specifications.


 * http://www.musl-libc.org/

the only way to truly appreciate musl is to look at the code and see what it's doing right - skarnet

sbase/ubase
This is a complete reimplementation of the GNU coreutils. The source code is shorter and tries to be simpler. They also fix a lot of oddities and bugs in details (like the tr program).


 * http://core.suckless.org/sbase
 * http://core.suckless.org/ubase

execline
A wild reinvention of shell based on taking chain loading to its logical conclusion.


 * http://skarnet.org/software/execline/

zenity
zenity is a very simple tool that just lets you create dialog boxes (of various kinds) from the command line. The code is bigger than it needs to be and it has a few MB of data (pictures, translations) but overall it's a good idea, has a small scope and as an abstraction it is quite well sealed off.


 * https://help.gnome.org/users/zenity/stable/

sortix libz
This libz has been cleaned up:

Support for anything but modern Unix operating systems has been deleted. Support for 16-bit segmented platforms has been deleted. K&R function prototypes and declarations have been upgraded to ISO C ones. Mysterious work-arounds for broken toolchains and operating systems have been deleted. and more


 * https://sortix.org/libz/

qbe
"QBE aims to be a pure C embeddable backend that provides 70% of the performance of advanced compilers in 10% of the code. Its small size serves both its aspirations of correctness and our ability to understand, fix, and improve it. It also serves its users by providing trivial integration and great flexibility."


 * http://c9x.me/compile/