diff --git a/README.md b/README.md
index 7cf5d16..7298175 100644
--- a/README.md
+++ b/README.md
@@ -69,8 +69,8 @@ firefox OCR-D-OCR-CALAMARI-EVAL/OCR-D-OCR-CALAMARI-EVAL_00000024.html
 
 ppn2ocr
 -------
-The `ppn2ocr` script produces a METS file with the best images for a given
-document in the State Library Berlin (SBB)'s digitized collection. 
+The `ppn2ocr` script produces a workspace and METS file with the best images for
+a given document in the State Library Berlin (SBB)'s digitized collection.
 
 Install it with an up-to-date pip (otherwise this will fail due to [a opencv-python-headless build failure](https://github.com/skvark/opencv-python#frequently-asked-questions)):
 ~~~
@@ -87,7 +87,19 @@ cd PPN77164308X
 This produces a workspace directory `PPN77164308X` with the OCR results in it;
 the results are viewable as explained above.
 
-ppn2ocr requires a working Docker setup and properly set up environment
-variables for the proxy configuration. At SBB, please read
-`howto/docker-proxy.md` and `howto/proxy-settings-for-shell+python.md`
-(in qurator's mono-repo).
+ppn2ocr requires properly set up environment variables for the proxy
+configuration. At SBB, please read `howto/docker-proxy.md` and
+`howto/proxy-settings-for-shell+python.md` (in qurator's mono-repo).
+
+ocrd-workspace-from-images
+--------------------------
+The `ocrd-workspace-from-images` script produces a OCR-D workspace (incl. METS)
+for the given images.
+
+~~~
+~/devel/my_ocrd_workflow/ocrd-workspace-from-images 0005.png
+cd workspace-sj4EH
+~/devel/my_ocrd_workflow/run-docker-hub -I BEST --skip-validation
+~~~
+
+This produces a workspace from the files and then runs the OCR workflow on it.
diff --git a/ocrd-workspace-from-images b/ocrd-workspace-from-images
new file mode 100755
index 0000000..919e454
--- /dev/null
+++ b/ocrd-workspace-from-images
@@ -0,0 +1,38 @@
+#!/bin/bash
+# Create an OCR-D workspace from images
+#
+#   ocrd-workspace-from-images *.png
+#
+# In order to produce a workspace that validates, this script makes best effort
+# to generate random IDs and to create the necessary structures like the
+# physical page sequence.
+
+workspace_dir=`mktemp -d "workspace-XXXXX"`
+workspace_id=`basename $workspace_dir`
+
+ocrd workspace -d $workspace_dir init
+ocrd workspace -d $workspace_dir set-id $workspace_id
+
+make_file_id_from_filename() {
+  filename="$1"
+  file_id="$filename"
+  file_id=`echo $file_id | sed 's#(.png|.tif|.jpe?g)$##i'`
+  file_id=`echo $file_id | sed 's#[^A-Za-z0-9_-]#_#g'`
+
+  echo "$file_id"
+}
+
+mkdir $workspace_dir/OCR-D-IMG
+page_count=0
+for img_orig in "$@"; do
+  page_count=$(($page_count + 1))
+  img="$workspace_dir/OCR-D-IMG/`basename $img_orig`"
+  cp -L "$img_orig" "$img"
+  file_id=`make_file_id_from_filename "$img"`
+  mime_type=`file -b --mime-type "$img"`
+  page_id=`printf "P%05d" $page_count`
+  ocrd workspace -d $workspace_dir add -G OCR-D-IMG "$img" --file-id $file_id --page-id $page_id --mimetype $mime_type
+done
+
+ocrd workspace -d $workspace_dir validate
+echo $workspace_dir