diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
index 7956e48..78a1940 100644
--- a/.github/workflows/test.yml
+++ b/.github/workflows/test.yml
@@ -123,3 +123,6 @@ jobs:
           echo "This is a schedule event"
           poetry run pytest -vv tests/
           poetry run pytest --openai_model=gpt-4o -m chat -vv tests/test_openai.py
+
+      - name: Test cli commands
+        run: poetry run pytest -vv tests/test_cli.py
\ No newline at end of file
diff --git a/log10/__main__.py b/log10/__main__.py
index 32409c5..07c9387 100644
--- a/log10/__main__.py
+++ b/log10/__main__.py
@@ -1,74 +1,5 @@
-try:
-    import click
-except ImportError:
-    print(
-        "To use log10 cli you must install optional modules. Please install them with `pip install 'log10-io[cliModules]'`."
-    )
-    exit(1)
+from log10.cli.cli_commands import cli
 
-from log10.cli.autofeedback import auto_feedback_icl, get_autofeedback_cli
-from log10.cli.completions import benchmark_models, download_completions, get_completion, list_completions
-from log10.cli.feedback import create_feedback, download_feedback, get_feedback, list_feedback
-from log10.cli.feedback_task import create_feedback_task, get_feedback_task, list_feedback_task
-
-
-@click.group()
-def cli():
-    pass
-
-
-@click.group()
-def completions():
-    """
-    Manage logs from completions i.e. logs from users
-    """
-    pass
-
-
-@click.group(name="feedback")
-def feedback():
-    """
-    Manage feedback for completions i.e. capturing feedback from users
-    """
-    pass
-
-
-@click.group(name="auto_feedback")
-def auto_feedback():
-    """
-    Manage auto feedback for completions i.e. capturing feedback from users
-    """
-    pass
-
-
-@click.group()
-def feedback_task():
-    """
-    Manage tasks for feedback i.e. instructions and schema for feedback
-    """
-    pass
-
-
-cli.add_command(completions)
-completions.add_command(list_completions, "list")
-completions.add_command(get_completion, "get")
-completions.add_command(download_completions, "download")
-completions.add_command(benchmark_models, "benchmark_models")
-
-cli.add_command(feedback)
-feedback.add_command(create_feedback, "create")
-feedback.add_command(list_feedback, "list")
-feedback.add_command(get_feedback, "get")
-feedback.add_command(download_feedback, "download")
-feedback.add_command(auto_feedback_icl, "predict")
-feedback.add_command(auto_feedback, "autofeedback")
-# Subcommands for auto_feedback under feedback command
-auto_feedback.add_command(get_autofeedback_cli, "get")
-
-cli.add_command(feedback_task)
-feedback_task.add_command(create_feedback_task, "create")
-feedback_task.add_command(list_feedback_task, "list")
-feedback_task.add_command(get_feedback_task, "get")
 
 if __name__ == "__main__":
     cli()
diff --git a/log10/cli/cli_commands.py b/log10/cli/cli_commands.py
new file mode 100644
index 0000000..2aa5685
--- /dev/null
+++ b/log10/cli/cli_commands.py
@@ -0,0 +1,71 @@
+try:
+    import click
+except ImportError:
+    print(
+        "To use log10 cli you must install optional modules. Please install them with `pip install 'log10-io[cliModules]'`."
+    )
+    exit(1)
+
+from log10.cli.autofeedback import auto_feedback_icl, get_autofeedback_cli
+from log10.cli.completions import benchmark_models, download_completions, get_completion, list_completions
+from log10.cli.feedback import create_feedback, download_feedback, get_feedback, list_feedback
+from log10.cli.feedback_task import create_feedback_task, get_feedback_task, list_feedback_task
+
+
+@click.group()
+def cli():
+    pass
+
+
+@click.group()
+def completions():
+    """
+    Manage logs from completions i.e. logs from users
+    """
+    pass
+
+
+@click.group(name="feedback")
+def feedback():
+    """
+    Manage feedback for completions i.e. capturing feedback from users
+    """
+    pass
+
+
+@click.group(name="auto_feedback")
+def auto_feedback():
+    """
+    Manage auto feedback for completions i.e. capturing feedback from users
+    """
+    pass
+
+
+@click.group()
+def feedback_task():
+    """
+    Manage tasks for feedback i.e. instructions and schema for feedback
+    """
+    pass
+
+
+cli.add_command(completions)
+completions.add_command(list_completions, "list")
+completions.add_command(get_completion, "get")
+completions.add_command(download_completions, "download")
+completions.add_command(benchmark_models, "benchmark_models")
+
+cli.add_command(feedback)
+feedback.add_command(create_feedback, "create")
+feedback.add_command(list_feedback, "list")
+feedback.add_command(get_feedback, "get")
+feedback.add_command(download_feedback, "download")
+feedback.add_command(auto_feedback_icl, "predict")
+feedback.add_command(auto_feedback, "autofeedback")
+# Subcommands for auto_feedback under feedback command
+auto_feedback.add_command(get_autofeedback_cli, "get")
+
+cli.add_command(feedback_task)
+feedback_task.add_command(create_feedback_task, "create")
+feedback_task.add_command(list_feedback_task, "list")
+feedback_task.add_command(get_feedback_task, "get")
diff --git a/tests/test_cli.py b/tests/test_cli.py
new file mode 100644
index 0000000..289054a
--- /dev/null
+++ b/tests/test_cli.py
@@ -0,0 +1,66 @@
+import pytest
+from click.testing import CliRunner
+
+from log10.cli.cli_commands import cli
+
+
+completion_id = "fb891d4e-f99c-4d8d-a95c-24d2ed7a0807"
+feedback_id = "0e43b537-5f0c-4f47-ba83-4938514477c3"
+feedback_task_id = "1c84079e-a7bb-47e1-86c7-e32b51045e8e"
+
+
+@pytest.fixture
+def runner():
+    return CliRunner()
+
+
+def test_list_completions(runner):
+    result = runner.invoke(cli, ["completions", "list"])
+    print(result.output)
+    assert result.exit_code == 0
+    assert "total_completions=" in result.output
+
+
+def test_get_completion(runner):
+    result = runner.invoke(cli, ["completions", "get", "--id", completion_id])
+    assert result.exit_code == 0
+    assert completion_id in result.output
+
+
+def test_download_completions(runner):
+    result = runner.invoke(cli, ["completions", "download", "--limit", "1", "--tags", "log10/summary-grading"])
+    assert result.exit_code == 0
+    assert "Download total completions: 1/" in result.output
+
+
+def test_list_feedback(runner):
+    result = runner.invoke(cli, ["feedback", "list"])
+    assert result.exit_code == 0
+    assert "Total feedback:" in result.output
+
+
+def test_get_feedback(runner):
+    result = runner.invoke(cli, ["feedback", "get", "--id", feedback_id])
+    assert result.exit_code == 0
+    assert feedback_id in result.output
+
+
+def test_download_feedback(runner):
+    result = runner.invoke(cli, ["feedback", "download", "--limit", "1"])
+    assert result.exit_code == 0
+
+
+def test_get_autofeedback(runner):
+    result = runner.invoke(cli, ["feedback", "autofeedback", "get", "--completion-id", completion_id])
+    assert result.exit_code == 0
+    assert completion_id in result.output
+
+
+def test_list_feedback_task(runner):
+    result = runner.invoke(cli, ["feedback-task", "list"])
+    assert result.exit_code == 0
+
+
+def test_get_feedback_task(runner):
+    result = runner.invoke(cli, ["feedback-task", "get", "--id", feedback_task_id])
+    assert result.exit_code == 0