diff --git a/.github/workflows/ec2-integration-test.yml b/.github/workflows/ec2-integration-test.yml
index 8d34cb935b..5df4f24b1c 100644
--- a/.github/workflows/ec2-integration-test.yml
+++ b/.github/workflows/ec2-integration-test.yml
@@ -56,6 +56,10 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data;
+      # account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(inputs.test_props) }}
     permissions:
diff --git a/.github/workflows/test-artifacts.yml b/.github/workflows/test-artifacts.yml
index 4c20eca962..e3c50363ce 100644
--- a/.github/workflows/test-artifacts.yml
+++ b/.github/workflows/test-artifacts.yml
@@ -449,6 +449,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ec2_gpu_matrix) }}
     permissions:
@@ -805,6 +808,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ec2_windows_matrix) }}
     permissions:
@@ -915,6 +921,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ec2_mac_matrix) }}
     permissions:
@@ -1070,6 +1079,9 @@ jobs:
     needs: [ GenerateTestMatrix, OutputEnvVariables ]
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ecs_ec2_launch_daemon_matrix) }}
     permissions:
@@ -1167,6 +1179,9 @@ jobs:
     needs: [ GenerateTestMatrix, OutputEnvVariables ]
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ecs_fargate_matrix) }}
     permissions:
@@ -1258,6 +1273,9 @@ jobs:
     needs: [ GenerateTestMatrix, OutputEnvVariables ]
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.eks_daemon_matrix) }}
     permissions:
@@ -1375,6 +1393,9 @@ jobs:
     needs: [ GenerateTestMatrix, OutputEnvVariables ]
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.eks_deployment_matrix) }}
     permissions:
@@ -1467,6 +1488,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ec2_performance_matrix) }}
     permissions:
@@ -1556,6 +1580,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ec2_windows_performance_matrix) }}
     permissions:
@@ -1645,6 +1672,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ec2_stress_matrix) }}
     permissions:
@@ -1738,6 +1768,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.ec2_windows_stress_matrix) }}
     permissions:
@@ -1833,6 +1866,9 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
+      # Cap concurrent jobs per matrix to limit simultaneous STS AssumeRoleWithWebIdentity calls.
+      # 100 is a conservative value, not calibrated against throttle data; account concurrency limits may cap lower.
+      max-parallel: 100
       matrix:
         arrays: ${{ fromJson(needs.GenerateTestMatrix.outputs.eks_addon_matrix) }}
     permissions: