diff --git a/README.md b/README.md
index 8c57771..cbf9a6d 100644
--- a/README.md
+++ b/README.md
@@ -11,7 +11,8 @@ Usage:
 - uses: mikeal/publish-to-github-action@master
   env:
     GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-    BRANCH_NAME: '' #optional defaults to master
+    BRANCH_NAME: '' #optional defaults to "master"
+    COMMIT_MESSAGE: '' #optional defaults to "Automated publish: ${timestamp} ${GITHUB_SHA}"
 ```
 
 If you can get away with only ever writing new files, instead of updating them, you won’t regret it ;) Every time you update a file you retain the history of the file. Over time this can become quite large and contain a lot of duplicate data, even when using Git LFS as described below. The only way to garbage collect this historical data is to use a [hefty Java program called BFG](https://rtyley.github.io/bfg-repo-cleaner/) to re-write the history.
diff --git a/entrypoint.sh b/entrypoint.sh
index 8139556..92154f4 100755
--- a/entrypoint.sh
+++ b/entrypoint.sh
@@ -10,6 +10,11 @@ if [ -z "${BRANCH_NAME}" ]; then
    export BRANCH_NAME=master
 fi
 
+if [ -z "${COMMIT_MESSAGE}" ]; then
+   timestamp=$(date -u)
+   export COMMIT_MESSAGE="Automated publish: ${timestamp} ${GITHUB_SHA}"
+fi
+
 # initialize git
 remote_repo="https://${GITHUB_ACTOR}:${GITHUB_TOKEN}@github.com/${GITHUB_REPOSITORY}.git"
 git config http.sslVerify false
@@ -25,7 +30,6 @@ git lfs install
 # publish any new files
 git checkout ${BRANCH_NAME}
 git add -A
-timestamp=$(date -u)
-git commit -m "Automated publish: ${timestamp} ${GITHUB_SHA}" || exit 0
+git commit -m "${COMMIT_MESSAGE}" || exit 0
 git pull --rebase publisher ${BRANCH_NAME}
 git push publisher ${BRANCH_NAME}