diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml
index 28058e3..1b540a7 100644
--- a/.github/workflows/python-package.yml
+++ b/.github/workflows/python-package.yml
@@ -17,7 +17,7 @@ jobs:
     steps:
     - uses: actions/checkout@v4
     - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v4
+      uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install dependencies
diff --git a/README.md b/README.md
index 0ed99da..a217b37 100644
--- a/README.md
+++ b/README.md
@@ -59,8 +59,9 @@ cldfbench zenodo --communities dplace cldfbench_<id>.py
 cldfbench cldfviz.map cldf --pacific-centered --format png --width 20 --output map.png --with-ocean --no-legend
 cldfbench readme cldfbench_<id>.py
 dplace check cldfbench_<id>.py
-git commit -a -m"release vX.Y"
+git commit -a -m"release v3.0"
 git push origin
+dplace release cldfbench_<id>.py v3.0
 ```
 
 Then create a release on GitHub, thereby pushing the repos to Zenodo.
diff --git a/src/pydplace/commands/release.py b/src/pydplace/commands/release.py
new file mode 100644
index 0000000..5f8733b
--- /dev/null
+++ b/src/pydplace/commands/release.py
@@ -0,0 +1,42 @@
+"""
+
+"""
+from cldfbench.cli_util import add_dataset_spec, get_dataset
+
+
+def register(parser):
+    add_dataset_spec(parser)
+    parser.add_argument('tag')
+
+
+def run(args):  # pragma: no cover
+    tag = args.tag
+    if not tag.startswith('v'):
+        tag = 'v' + tag
+    ds = get_dataset(args)
+    cldf = ds.cldf_reader()
+    if cldf.properties.get('dc:bibliographicCitation'):
+        relnotes = """\
+Cite the source as
+
+> {}
+
+and the D-PLACE dataset as
+
+DOI""".format(cldf.properties['dc:bibliographicCitation'])
+    else:
+        relnotes = """\
+Cite the D-PLACE dataset as
+
+DOI"""
+
+    ds.dir.joinpath('relnotes.txt').write_text(relnotes, encoding='utf8')
+    print('gh release create {} --title "{}" --notes-file relnotes.txt'.format(
+        tag, cldf.properties['dc:title']))
+    print('git pull origin')
+    print('')
+    print("Now you should grab the Zenodo version DOI from\n"
+          "https://zenodo.org/account/settings/github/repository/D-PLACE/{0}\n"
+          "and add it to\n"
+          "https://github.com/D-PLACE/{0}/releases/edit/{1}\n and the concept DOI to "
+          "metdata.json".format(cldf.properties['rdf:ID'], tag))
diff --git a/src/pydplace/dataset.py b/src/pydplace/dataset.py
index 01de53d..cc44e6e 100644
--- a/src/pydplace/dataset.py
+++ b/src/pydplace/dataset.py
@@ -285,6 +285,12 @@ def add_society(self, writer, **props):
         )
         writer.objects['LanguageTable'].append(props)
 
+    def local_makecldf(self, args):
+        """
+        Datasets can hook up any custom processing to happen at the end of the CLDF conversion.
+        """
+        pass
+
     def cmd_makecldf(self, args):
         data_schema(args.writer.cldf)
         self.schema(args.writer.cldf)
@@ -309,6 +315,7 @@ def cmd_makecldf(self, args):
                 comment=row['Comment'],
                 glottocode_comment=row['glottocode_comment'],
             )
+        self.local_makecldf(args)
 
     def cmd_readme(self, args):
         print('gh repo edit --description "{}" --add-topic "ethnology"'.format(self.metadata.title))