diff --git a/cyberai/cli/bench.py b/cyberai/cli/bench.py
index 06ed75e..f0c0365 100644
--- a/cyberai/cli/bench.py
+++ b/cyberai/cli/bench.py
@@ -166,10 +166,12 @@ def _model_participation(report) -> tuple[int | None, str | None]:
     a value.
     """
     results = list(report.results)
-    if not results:
-        return None, None
     proven = [r for r in results if r.details.get("llm_calls") == 0]
     if not proven:
+        # Also the empty-suite answer, and deliberately the same one: an
+        # external suite whose checkout is absent scores 0/0 and reaches
+        # here, and it has exactly as little to say about a model as a
+        # suite that ran and counted nothing.
         return None, None
     if len(proven) == len(results):
         reasons = {str(r.details.get("llm_zero_reason")) for r in proven}
diff --git a/tests/unit/test_bench_cli.py b/tests/unit/test_bench_cli.py
index cf2df21..09ff399 100644
--- a/tests/unit/test_bench_cli.py
+++ b/tests/unit/test_bench_cli.py
@@ -490,3 +490,21 @@ def test_the_placeholder_engine_publishes_no_model_row(tmp_path):
     text = out.read_text()
     assert "| llm calls |" not in text
     assert "| llm zero reason |" not in text
+
+
+def test_a_suite_with_no_tasks_writes_a_card_and_claims_nothing(monkeypatch, tmp_path):
+    """An external suite whose checkout is absent scores 0/0 and still
+    writes a card. Nothing ran, so nothing can be said about a model --
+    and the rollup must answer that rather than divide by an empty run."""
+    monkeypatch.setenv("CVEBENCH_DIR", str(tmp_path / "nowhere"))
+    out = tmp_path / "sc.md"
+
+    result = CliRunner().invoke(
+        bench, ["run", "--suite", "cve-bench", "--engine", "agent", "--scorecard", str(out)]
+    )
+
+    assert result.exit_code == 0
+    assert "pass@1: 0/0" in result.output
+    text = out.read_text()
+    assert "| llm calls |" not in text
+    assert "| llm zero reason |" not in text