diff --git a/.github/workflows/e2e-tests.yml b/.github/workflows/e2e-tests.yml
index 63e0d3162..22d4304e7 100644
--- a/.github/workflows/e2e-tests.yml
+++ b/.github/workflows/e2e-tests.yml
@@ -261,6 +261,7 @@ jobs:
                   MULTI_AGENTIC_EVAL=$(echo "$CONFIG_JSON" | python3 -c "import sys,json; d=json.load(sys.stdin); print(json.dumps([x for x in d if x.get('scenario-type') == 'agentic-coding' and 'prefill' in x and x.get('run-eval', False)]))" | score_matrix multi-agentic-eval)
                   SINGLE=$(echo "$CONFIG_JSON" | python3 -c "import sys,json; d=json.load(sys.stdin); print(json.dumps([x for x in d if 'prefill' not in x and x.get('scenario-type') != 'agentic-coding' and not x.get('eval-only', False)]))" | score_matrix single)
                   MULTI=$(echo "$CONFIG_JSON" | python3 -c "import sys,json; d=json.load(sys.stdin); print(json.dumps([x for x in d if 'prefill' in x and x.get('scenario-type') != 'agentic-coding' and not x.get('eval-only', False)]))" | score_matrix multi)
+                  EVALS=$(echo "$CONFIG_JSON" | python3 -c "import sys,json; d=json.load(sys.stdin); print(json.dumps([x for x in d if 'prefill' not in x and x.get('scenario-type') != 'agentic-coding' and x.get('run-eval', False)]))" | score_matrix eval)
                   MULTI_EVAL=$(echo "$CONFIG_JSON" | python3 -c "import sys,json; d=json.load(sys.stdin); print(json.dumps([x for x in d if 'prefill' in x and x.get('scenario-type') != 'agentic-coding' and x.get('run-eval', False)]))" | score_matrix multi-eval)
                   {
                     echo "agentic-config=$AGENTIC"
@@ -275,7 +276,7 @@ jobs:
 
     test-sweep-multi-node:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.multi-node-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.multi-node-config != '' && needs.get-jobs.outputs.multi-node-config != '[]' }}
         uses: ./.github/workflows/benchmark-multinode-tmpl.yml
         name: multi-node /
         strategy:
@@ -330,7 +331,7 @@ jobs:
 
     test-sweep-multi-node-evals:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.multi-node-eval-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.multi-node-eval-config != '' && needs.get-jobs.outputs.multi-node-eval-config != '[]' }}
         uses: ./.github/workflows/benchmark-multinode-tmpl.yml
         name: multi-node eval /
         strategy:
@@ -385,7 +386,7 @@ jobs:
 
     test-sweep-agentic:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.agentic-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.agentic-config != '' && needs.get-jobs.outputs.agentic-config != '[]' }}
         uses: ./.github/workflows/benchmark-tmpl.yml
         name: agentic /
         strategy:
@@ -431,7 +432,7 @@ jobs:
 
     test-sweep-agentic-evals:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.agentic-eval-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.agentic-eval-config != '' && needs.get-jobs.outputs.agentic-eval-config != '[]' }}
         uses: ./.github/workflows/benchmark-tmpl.yml
         name: agentic eval /
         strategy:
@@ -473,7 +474,7 @@ jobs:
 
     test-sweep-multi-node-agentic:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.multi-node-agentic-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.multi-node-agentic-config != '' && needs.get-jobs.outputs.multi-node-agentic-config != '[]' }}
         uses: ./.github/workflows/benchmark-multinode-tmpl.yml
         name: multi-node agentic /
         strategy:
@@ -533,7 +534,7 @@ jobs:
 
     test-sweep-multi-node-agentic-evals:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.multi-node-agentic-eval-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.multi-node-agentic-eval-config != '' && needs.get-jobs.outputs.multi-node-agentic-eval-config != '[]' }}
         uses: ./.github/workflows/benchmark-multinode-tmpl.yml
         name: multi-node agentic eval /
         strategy:
@@ -594,7 +595,7 @@ jobs:
 
     test-sweep-single-node:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.single-node-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.single-node-config != '' && needs.get-jobs.outputs.single-node-config != '[]' }}
         uses: ./.github/workflows/benchmark-tmpl.yml
         name: single-node /
         strategy:
@@ -633,7 +634,7 @@ jobs:
 
     test-sweep-evals:
         needs: get-jobs
-        if: ${{ needs.get-jobs.outputs.eval-config != '[]' }}
+        if: ${{ needs.get-jobs.outputs.eval-config != '' && needs.get-jobs.outputs.eval-config != '[]' }}
         uses: ./.github/workflows/benchmark-tmpl.yml
         name: eval /
         strategy:
diff --git a/utils/test_ci_priority.py b/utils/test_ci_priority.py
index 1bef5fcda..9fa61a2a1 100644
--- a/utils/test_ci_priority.py
+++ b/utils/test_ci_priority.py
@@ -1,4 +1,5 @@
 import json
+import re
 import shlex
 from copy import deepcopy
 from decimal import Decimal
@@ -260,3 +261,17 @@ def test_queue_tokens_change_between_run_attempts():
         "123:2",
         ("0",),
     )
+
+
+def test_e2e_matrix_outputs_reference_initialized_shell_variables():
+    workflow = (
+        Path(__file__).parents[1] / ".github" / "workflows" / "e2e-tests.yml"
+    ).read_text()
+    assignments = set(
+        re.findall(r"^\s*([A-Z][A-Z0-9_]*)=\$\(", workflow, re.MULTILINE)
+    )
+    output_variables = set(
+        re.findall(r'^\s*echo "[a-z-]+=\$([A-Z][A-Z0-9_]*)"', workflow, re.MULTILINE)
+    )
+
+    assert output_variables <= assignments