My config and setup tips for my future self
Fish: Fix prompt for fish v3.4.1
Scripts: Add yadm install/upgrade script
Scripts: Forgot to add nvm install script


browse  log 



You can also use your local clone with git send-email.


Welcome to my personal collection of weirdish—sometimes unmaintained—configuration and lose organization of cool scripts!

These are the setup I have on almost all machines I work in, and because I work on many tildes, I make this set of configuration as cross-platform as possible.

Mirrors: tildegit (gitea) | GitHub

Table of Contents



These are more like "what I tend to do" in this repo

  • Mostly bash shebangs
  • Shared aliases and env between shells
  • Setup and install scripts
  • Private configuration tracked with yadm


  1. Install yadm
  2. Clone the repo with yadm
  3. Fix conflicts as you please
  4. Reload sessions
  • fish
  • tmux source ~/.tmux.conf
  1. Selectively run scripts in dotscripts using dot <type> <target> such as dot setup fish


Here are detailed information for each compenent of my dotfiles


I don't track bashrc or bash_profile because I like to keep it to the system's defaults.

The fish shell configuration is at .config/fish/ and config.fish doesn't really anything specific, it just sources the shared environment variable file (.exportenvs.fish) and the shared aliases (.aliases).

The .exportenvs.fish file is generated by dotscripts/gen/fish-exportenvs. It takes .exportenvs and translates it into fish syntax. Instead of using bass to source .exportenvs on the fly during fish's startup, fish can source the generated .exportenvs.fish directly which improves performance signifanctly.

I have a symlink .bash_aliases pointing to .aliases because bash likes to look for that file.

.exportenvs is basically a bunch of environment variables exports. A whole ton of installation scripts on the internet likes to add export something=something to bashrc, so when that happens I tend to just move it into my .exportenvs.

#oh my fish

oh my fish is like a plugin manager for fish, I don't have a lot of plugins, just these utilities:

  • z: quickly access a common dir
  • bass: source bash scripts and expressions in fish (I use this for sourcing .exportenvs)
  • pj: access projects

fish theme is based entirely on the terminal and the prompt is copied from a particular pre-existing prompt style that shows error status and git status.

#fish functions

fish functions (located at .config/fish/functions) are mostly aliases that require some checking or additional logic.

cat, rm, ll are aliased to ccat, trash, and exa respectively if those programs are installed.


The .config/fish/config_local.fish file is for configuration specific for a computer, this could be setting a special $EDITOR, etc.

Setup: dotscripts/setup/fish


  • vim: only for systems that don't have neovim installed
  • nvim: I use this as my primary editor and IDE
  • emacs: just to play around and learn elisp
  • doom: looking for IDE features and inspiration for my nvim setup. (Currenting editting README using doom). This may become my primary IDE for GUI.
  • vscode: I only use this when I'm too stressed to remember vim/emacs's key binds, or sometimes when I'm remote-editting my site's markdown posts.
#vim and nvim

The bin/nv script is an alias to neovim, and runs vim if neovim is not installed.

For vim I use Vundle as my plugin manager just because I discovered vim before nvim and Vundle before Plug. My vimrc is largely unmaintained pieces of configuration copied from other people's configs. The set of plugins are mostly maintained (as in me modifying what plugins to use), except for auto-pairs if I remember correctly.

Sometimes when I change a plugin for nvim and change it for vim too, if I have extra time.

I use vim-plug as my plugin manager for nvim because it is shorter to type. Completion with CoC is set up for languages that I commonly work in. CoC keybinds configuration is mostly copied from their readme.

My nvim config directory (.config/nvim) has separate files for different types of configuration. general.vim has mostly the same content as .vimrc.

The isWSL function checks whether the system is WSL in my (n)vim configs I rebind some keys to make it work on WSL, but I don't use it much anymore.

Setup: dotscripts/setup/nvim


I use chemacs2 for switching emacs profiles and it allows me to use both vanilla emacs and doom emacs at the same time. See .emacs-profiles.el.

Setup: dotscripts/setup/emacs. This script is mostly tested and it should set up everything you need to start using both emacs and doom normally. It installs chemacs and doom if not already.

#Vanilla emacs

They should probably work for version 26 too, but I use 27+ on almost all my machines.

Stuff in .config/emacsd are largely copied from various emacs distributions to suit my needs. The modules directory is work-in-progress.

#Doom emacs

Doom itself requires at least emacs 27+ and git 2.23+ among other things. More info on the github repo.

There isn't much personal configuration in .config/doom yet, it's mostly just enabling or disabling doom modules and add some packages.

The emacs directory would be at .doomemacs (because .emacs.d is for chemacs).

#Gemini and Spartan client

My amfora config at .config/amfora uses the default theme with mostly default settings. I don't use amfora all that often because I also use elpher, lagrange, and gelim. I plan to have a custom newtab page as well as a custom theme in the future. Because gelim doesn't support client certificates yet, I only mostly use amfora for capsules that want a client certificate.

gelim config is at .config/gelim. gelim is my own gemini and spartan client for the terminal that isn't a full-blown TUI and tries to stay simple whilst offering many features and extensibility. It's still WIP so there won't be much configuration in there yet.


  • [ ] add LSP and completion to emacs
  • [x] (n)vim mapping for gg= (not needed anymore since we had g@ mapping for LSP format)
  • [x] .addpath file