Generate git commit message from git-status. Will generate a commit message like "Added: file1.py file2.py file3.py Modified: file4.py file5.py Deleted: README.md Renamed: test.txt-> test2.txt". Put this in your .gitconfig.

..git-commit-status - Generate git commit message from git-status.md

git commit-status alias

An alias that will generate a git commit message staged changes as shown in git-status. Put this alias (section below) in your .gitconfig.

The message generated will be in the format of:

$ git status --porcelain
A file1.py
A file2.py
A file3.py
M file4.py
M file5.py
D README.md
R test.txt-> test2.txt
$ git commit-status
$ git log --no-decorate -n 1
bee4f8e Added: file1.py file2.py file3.py Modified: file4.py file5.py Deleted: README.md Renamed: test.txt-> test2.txt

Shared on Stack Overflow.

.gitconfig

[alias]
	# commit-status: generate a commit with message from git-status (staged changes).
	# Source: https://gist.github.com/erikw/654386d35ecfdb0354cd2b71763f19ae
	# Explanation:
	# - Get only staged changes
	# - Ignore changes in working area (2nd letter, the Y in XY as explained in $(git help status))
	# - + split label and file path to separate lines so we can process the labels separately
	# - Keep only the first label using awk
	# - Add newline before each label section so we later can truncate \n to put everything on one line
	# - Make labels human readable e.g. M -> Modified
	# - Put everything on one line and trim leading & trailing whitespaces
	commit-status = !" \
	        TMPFILE=$(mktemp /tmp/git-commit-status-message.XXX); \
		git status --porcelain \
		  | grep '^[MARCDT]' \
		  | sort \
		  | sed -re 's/^([[:upper:]])[[:upper:]]?[[:space:]]+/\\1:\\n/' \
		  | awk '!x[$0]++' \
		  | sed -re 's/^([[:upper:]]:)$/\\n\\1/' \
		  | sed -re 's/^M:$/Modified: /' \
		  | sed -re 's/^A:$/Added: /' \
		  | sed -re 's/^R:$/Renamed: /' \
		  | sed -re 's/^C:$/Copied: /' \
		  | sed -re 's/^D:$/Deleted: /' \
		  | sed -re 's/^T:$/File Type Changed: /' \
		  | tr '\n' ' ' | xargs \
		  > $TMPFILE; \
		git commit -F $TMPFILE; \
		rm -f $TMPFILE \
		"




	# If you want to have a [Yn] prompt showing the commit message before making the commit,
	# then simply inject the following lines below in to the alias above just before the
	# "git commit -F $TMPFILE; \" line:
		cat $TMPFILE; \
	        commit=''; \
	        while :; do \
			echo '> Commit with this message? [Yn]: '; \
			read commit; \
			([ -z \"$commit\" ] || [ \"$commit\" = y ] || [ \"$commit\" = Y ] || [ \"$commit\" = n ]) && break; \
	        done; \
		test \"$commit\" != n || exit; \

Hi everyone
I just wanted to mention that in Android Termux shell, this produces output like

1:npath/to/file1.md 1:npath/to/file2.txt

Versions it's using:

bash --version
GNU bash, version 5.2.26(1)-release (aarch64-unknown-linux-android)

git --version
git version 2.44.0

sed --version
sed (GNU sed) 4.9

grep --version
grep (GNU grep) 3.11
...
grep -P uses PCRE2 10.43 2024-02-16

awk --version
GNU Awk 5.3.0, API 4.0, (GNU MPFR 4.2.1, GNU MP 6.3.0)

tr --version
tr (GNU coreutils) 9.4

I've tried playing with the double antislashes since it seems to be an escaping issue but cannot get the desired output. Any idea?

I have modified the script so that the changed file names are surrounded by backticks:

[alias]
	# commit-status: generate a commit with message from git-status (staged changes).
	# Source: https://gist.github.com/erikw/654386d35ecfdb0354cd2b71763f19ae
	# Explanation:
	# - Get only staged changes
	# - Ignore changes in working area (2nd letter, the Y in XY as explained in $(git help status))
	# - Wrap the file names in backticks
	# - + split label and file path to separate lines so we can process the labels separately
	# - Keep only the first label using awk
	# - Add newline before each label section so we later can truncate \n to put everything on one line
	# - Make labels human readable e.g. M -> Modified
	commit-status = !" \
            TMPFILE=$(mktemp /tmp/git-commit-status-message.XXX); \
            git status --porcelain \
              | grep '^[MARCDT]' \
              | sort \
              | sed -re 's/^([MARCDT][[:space:]]+)(.*)$/\\1 \\`\\2\\`/' \
              | sed -re 's/^([[:upper:]])[[:upper:]]?[[:space:]]+/\\1:\\n/' \
              | awk '!x[$0]++' \
              | sed -re 's/^([[:upper:]]:)$/\\n\\1/' \
              | sed -re 's/^M:$/Modified: /' \
              | sed -re 's/^A:$/Added: /' \
              | sed -re 's/^R:$/Renamed: /' \
              | sed -re 's/^C:$/Copied: /' \
              | sed -re 's/^D:$/Deleted: /' \
              | sed -re 's/^T:$/File Type Changed: /' \
              > $TMPFILE; \
            git commit -F $TMPFILE; \
            rm -f $TMPFILE \
            "

Also removed the linebreak trimming, as I think this looks a bit cleaner. Which of course is highly subjective.
Thank you very much, @erikw, for creating and uploading this gist - and other contributors!