diff --git a/ampel/ztf/archive/server/s3.py b/ampel/ztf/archive/server/s3.py
index e2f3826..3024d27 100644
--- a/ampel/ztf/archive/server/s3.py
+++ b/ampel/ztf/archive/server/s3.py
@@ -1,6 +1,7 @@
 from functools import lru_cache
 import io
 from typing import TYPE_CHECKING
+from urllib.parse import urlsplit
 
 import boto3
 from botocore.exceptions import ClientError
@@ -13,8 +14,9 @@
     from mypy_boto3_s3.service_resource import Bucket
     from typing import BinaryIO
 
-class NoSuchKey(KeyError):
-    ...
+
+class NoSuchKey(KeyError): ...
+
 
 @lru_cache(maxsize=1)
 def get_s3_bucket() -> "Bucket":
@@ -38,7 +40,7 @@ def get_object(bucket: "Bucket", key: str) -> bytes:
 def get_stream(bucket: "Bucket", key: str) -> "BinaryIO":
     response = bucket.Object(key).get()
     if response["ResponseMetadata"]["HTTPStatusCode"] <= 400:
-        return response["Body"] # type: ignore[return-value]
+        return response["Body"]  # type: ignore[return-value]
     else:
         raise KeyError
 
@@ -60,10 +62,16 @@ def get_range(
             schema = get_parsed_schema(read_schema(get_stream(bucket, key)))
         else:
             schema = ALERT_SCHEMAS[schema_key]
-        return response["Body"], schema # type: ignore[return-value]
+        return response["Body"], schema  # type: ignore[return-value]
     else:
         raise KeyError
 
 
 def get_url_for_key(bucket: "Bucket", key: str) -> str:
     return f"{settings.s3_endpoint_url or ''}/{bucket.name}/{key}"
+
+
+def get_key_for_url(bucket: "Bucket", uri: str) -> str:
+    path = urlsplit(uri).path.split("/")
+    assert path[-2] == bucket.name
+    return path
diff --git a/migrations/02_add_archive_count.py b/migrations/02_add_archive_count.py
new file mode 100644
index 0000000..1869926
--- /dev/null
+++ b/migrations/02_add_archive_count.py
@@ -0,0 +1,48 @@
+#!/usr/bin/env python
+
+import logging
+import time
+from argparse import ArgumentParser
+import sqlalchemy as sa
+import fastavro
+
+import boto3
+
+from ampel.ztf.t0.ArchiveUpdater import ArchiveUpdater
+from ampel.ztf.archive.server.s3 import get_object, get_key_for_url
+
+
+logging.basicConfig(level="INFO", format="[%(asctime)s] %(message)s")
+log = logging.getLogger()
+
+parser = ArgumentParser()
+parser.add_argument("uri")
+parser.add_argument("--s3-bucket")
+parser.add_argument("--s3-endpoint")
+
+
+args = parser.parse_args()
+
+engine = sa.engine.create_engine(args.uri)
+
+bucket = boto3.resource("s3", endpoint_url=args.s3_endpoint).Bucket(args.s3_bucket)
+
+with engine.connect() as connection:
+    meta = sa.MetaData(connection)
+    meta.reflect()
+
+    Alert = meta.tables["alert"]
+    Archive = meta.tables["avro_archive"]
+
+    query = Archive.select(Archive.c.alerts is None)
+
+    for row in connection.execute(query).fetchone():
+        key = get_key_for_url(row["uri"])
+        reader = fastavro.reader(get_object(bucket, key))
+        alerts = len(list(reader))
+
+        print(row, alerts)
+
+        assert False
+
+        # TODO: update alerts field
diff --git a/tests/test-data/initdb/archive/25_avro_archive_count.sql b/tests/test-data/initdb/archive/25_avro_archive_count.sql
new file mode 100644
index 0000000..0539bc4
--- /dev/null
+++ b/tests/test-data/initdb/archive/25_avro_archive_count.sql
@@ -0,0 +1,5 @@
+BEGIN;
+
+ALTER TABLE avro_archive ADD count INTEGER;
+
+COMMIT;
\ No newline at end of file