diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index 73adf952..d6935e58 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -37,9 +37,12 @@ jobs:
           echo "--- JSON Results ---"
           cat /tmp/build-results.json
 
-          # Check if any tests failed
-          if grep -q '"pass": false' /tmp/build-results.json; then
-            echo "Some build tests failed"
+      - name: Check test results
+        run: |
+          FAILED=$(jq '.summary.failed' /tmp/build-results.json)
+          echo "Failed tests: $FAILED"
+          if [ "$FAILED" -gt 0 ]; then
+            echo "::error::$FAILED build test(s) failed"
             exit 1
           fi
 
diff --git a/.github/workflows/full-pipeline.yml b/.github/workflows/full-pipeline.yml
index 81c92d3d..a29dcf13 100644
--- a/.github/workflows/full-pipeline.yml
+++ b/.github/workflows/full-pipeline.yml
@@ -67,6 +67,15 @@ jobs:
           echo "--- JSON Results ---"
           cat /tmp/runtime-results.json
 
+      - name: Check test results
+        run: |
+          FAILED=$(jq '.summary.failed' /tmp/runtime-results.json)
+          echo "Failed tests: $FAILED"
+          if [ "$FAILED" -gt 0 ]; then
+            echo "::error::$FAILED runtime test(s) failed"
+            exit 1
+          fi
+
       - name: Upload runtime results
         uses: actions/upload-artifact@v4
         if: always()
@@ -100,6 +109,15 @@ jobs:
           echo "--- JSON Results ---"
           cat /tmp/inference-results.json
 
+      - name: Check test results
+        run: |
+          FAILED=$(jq '.summary.failed' /tmp/inference-results.json)
+          echo "Failed tests: $FAILED"
+          if [ "$FAILED" -gt 0 ]; then
+            echo "::error::$FAILED inference test(s) failed"
+            exit 1
+          fi
+
       - name: Upload inference results
         uses: actions/upload-artifact@v4
         if: always()
@@ -139,6 +157,15 @@ jobs:
           echo "--- JSON Results ---"
           cat /tmp/llm-judged-results.json
 
+      - name: Check test results
+        run: |
+          FAILED=$(jq '.summary.failed' /tmp/llm-judged-results.json)
+          echo "Failed tests: $FAILED"
+          if [ "$FAILED" -gt 0 ]; then
+            echo "::error::$FAILED test(s) failed LLM evaluation"
+            exit 1
+          fi
+
       - name: Upload final results
         uses: actions/upload-artifact@v4
         if: always()
diff --git a/.github/workflows/inference.yml b/.github/workflows/inference.yml
index 173bad8d..f1b5de43 100644
--- a/.github/workflows/inference.yml
+++ b/.github/workflows/inference.yml
@@ -63,6 +63,15 @@ jobs:
           echo "--- JSON Results ---"
           cat /tmp/inference-results.json
 
+      - name: Check test results
+        run: |
+          FAILED=$(jq '.summary.failed' /tmp/inference-results.json)
+          echo "Failed tests: $FAILED"
+          if [ "$FAILED" -gt 0 ]; then
+            echo "::error::$FAILED inference test(s) failed"
+            exit 1
+          fi
+
       - name: Upload inference results
         uses: actions/upload-artifact@v4
         if: always()
diff --git a/.github/workflows/runtime.yml b/.github/workflows/runtime.yml
index fde739c3..70b863f0 100644
--- a/.github/workflows/runtime.yml
+++ b/.github/workflows/runtime.yml
@@ -59,6 +59,15 @@ jobs:
           echo "--- JSON Results ---"
           cat /tmp/runtime-results.json
 
+      - name: Check test results
+        run: |
+          FAILED=$(jq '.summary.failed' /tmp/runtime-results.json)
+          echo "Failed tests: $FAILED"
+          if [ "$FAILED" -gt 0 ]; then
+            echo "::error::$FAILED runtime test(s) failed"
+            exit 1
+          fi
+
       - name: Upload runtime results
         uses: actions/upload-artifact@v4
         if: always()