Git Daily Workflow

With the repository, working directory, and staging area model in place, this section focuses on the group of commands used most often day to day: recording changes, viewing history, and working with branches.

Initializing and checking status

git init                 # turn the current directory into a Git repository
git status                # check the current state of the working directory and staging area

git status is one of the most frequently run commands — it lists which files have been modified, which are already staged, and which are new files Git isn’t tracking yet. It’s the basic way to confirm the current state before doing anything else.

Recording a change

git add <file>: stage the current changes in a specific file; git add . stages every change in the current directory at once
git diff: view the specific changes in the working directory relative to the staging area (or the staging area relative to the last commit) — a way to confirm the scope of a change before staging it
git commit -m "commit message": write the staged content into the repository, creating a new commit

The commit message isn’t a throwaway comment — it’s part of the permanent history. Tracing when and why a particular change was introduced relies mainly on these messages, so it’s worth writing something concise but specific about what the change actually does, rather than a vague “update” or “fix.”

Viewing history

git log                  # view commit history, newest first
git log --oneline        # collapse each commit to one line, showing just the hash prefix and message
git log -p               # show the specific changes (diff) for each commit

Branches: parallel sequences of commits

A branch is, at its core, just a movable pointer to a commit — not a complete, independent copy of the files. Creating a new branch is nearly free, because it only creates a new pointer without copying any file content — which is why Git’s branching is noticeably lighter-weight than in many other version control tools.

git branch                    # list all current branches
git branch new-feature        # create a new branch without switching to it
git checkout new-feature      # switch to the given branch
git checkout -b new-feature   # create and immediately switch to a new branch (combines the two commands above)
git switch new-feature        # a newer, more purpose-specific command for switching branches

HEAD is a special internal pointer Git uses to mark “where things currently are.” Normally it points at the current branch, and the current branch’s pointer in turn points at that branch’s latest commit. Switching branches is, under the hood, just moving HEAD to point at a different branch.

Merging branches

git checkout main        # switch back to the target branch first (usually main)
git merge new-feature    # merge new-feature's commits into the current branch

If two branches modify the same part of the same file differently, git merge can’t automatically decide which change to keep, producing a merge conflict. Git marks each side’s version in the conflicting file with special markers (<<<<<<<, =======, >>>>>>>), and the file needs to be edited by hand to settle on a final version, then re-git added and git commited to complete the merge.