test(e2e): finish sandbox image suite migration

.github/workflows/e2e.yaml

@@ -4222,7 +4222,7 @@ jobs:
       E2E_AGENT_RUNTIME: "openclaw + hermes + langchain-deepagents-code"
       E2E_OBSERVABLE_OUTCOME: "Exact managed images start directly on the native architecture"
       RELEASE_E2E_ACTIVATION_PATH: ci/protected-managed-image-multiarch-activation-v1.json
-      NEMOCLAW_E2E_EXPECTED_SHA: ${{ inputs.checkout_sha }}
+      NEMOCLAW_E2E_EXPECTED_SHA: ${{ inputs.checkout_sha || github.sha }}
       NEMOCLAW_PROTECTED_MANAGED_IMAGE_HEAD_SHA: ${{ inputs.checkout_sha || github.sha }}
       NEMOCLAW_E2E_SHARD: ${{ matrix.shard }}
       NEMOCLAW_PROTECTED_MANAGED_IMAGE_BASE_SHA: ${{ inputs.base_sha || github.event.before || github.sha }}
@@ -4236,6 +4236,20 @@ jobs:
       NEMOCLAW_PROTECTED_REGISTRY_NAME: nemoclaw-managed-${{ matrix.shard }}-${{ github.run_id }}-${{ github.run_attempt }}
       NEMOCLAW_RUN_LIVE_E2E: "1"
     steps:
+      - name: Bind protected E2E correlation identity
+        shell: bash
+        env:
+          REQUESTED_CORRELATION_ID: ${{ inputs.correlation_id }}
+        run: |
+          set -euo pipefail
+          correlation_id="$REQUESTED_CORRELATION_ID"
+          if [[ -z "$correlation_id" ]]; then
+            correlation_id="$(node --input-type=module -e \
+              'import { randomUUID } from "node:crypto"; console.log(randomUUID())')"
+          fi
+          [[ "$correlation_id" =~ ^[a-f0-9]{8}-[a-f0-9]{4}-4[a-f0-9]{3}-[89ab][a-f0-9]{3}-[a-f0-9]{12}$ ]]
+          printf 'NEMOCLAW_E2E_CORRELATION_ID=%s\n' "$correlation_id" >>"$GITHUB_ENV"
+
       - name: Validate protected exact-head dispatch
         env:
           BASE_SHA: ${{ inputs.base_sha || github.event.before || github.sha }}
@@ -4444,7 +4458,7 @@ jobs:
           fi
           docker run --detach \
             --name "$NEMOCLAW_PROTECTED_REGISTRY_NAME" \
-            --label "io.nvidia.nemoclaw.e2e-owner=${NEMOCLAW_PROTECTED_MANAGED_IMAGE_COHORT}" \
+            --label "io.nvidia.nemoclaw.managed-image.cohort=${NEMOCLAW_PROTECTED_MANAGED_IMAGE_COHORT}" \
             --label "io.nvidia.nemoclaw.e2e-platform=${NEMOCLAW_PROTECTED_MANAGED_IMAGE_PLATFORM}" \
             --publish 127.0.0.1:5000:5000 \
             docker.io/library/registry@sha256:a3d8aaa63ed8681a604f1dea0aa03f100d5895b6a58ace528858a7b332415373
@@ -4540,29 +4554,23 @@ jobs:
             "${NEMOCLAW_PROTECTED_MANAGED_IMAGE_EVIDENCE}.tmp" \
             "$NEMOCLAW_PROTECTED_MANAGED_IMAGE_EVIDENCE"
 
+
       - name: Remove isolated protected managed-image registry
         if: always()
         shell: bash
         run: |
           set -euo pipefail
           if docker container inspect "$NEMOCLAW_PROTECTED_REGISTRY_NAME" >/dev/null 2>&1; then
-            owner="$(
-              docker container inspect \
-                --format '{{index .Config.Labels "io.nvidia.nemoclaw.e2e-owner"}}' \
-                "$NEMOCLAW_PROTECTED_REGISTRY_NAME"
-            )"
+            owner="$(docker container inspect --format '{{index .Config.Labels "io.nvidia.nemoclaw.managed-image.cohort"}}' "$NEMOCLAW_PROTECTED_REGISTRY_NAME")"
             [[ "$owner" == "$NEMOCLAW_PROTECTED_MANAGED_IMAGE_COHORT" ]] || {
               echo "::error::Refusing to remove a registry not owned by this protected shard" >&2
               exit 1
             }
             docker rm -f "$NEMOCLAW_PROTECTED_REGISTRY_NAME" >/dev/null
           fi
-          if docker container inspect "$NEMOCLAW_PROTECTED_REGISTRY_NAME" >/dev/null 2>&1; then
-            echo "::error::Protected managed-image registry container remained after cleanup" >&2
-            exit 1
-          fi
-          if curl --fail --silent --show-error http://127.0.0.1:5000/v2/ >/dev/null 2>&1; then
-            echo "::error::Protected managed-image registry listener remained after cleanup" >&2
+          if docker container inspect "$NEMOCLAW_PROTECTED_REGISTRY_NAME" >/dev/null 2>&1 ||
+            curl --fail --silent --show-error http://127.0.0.1:5000/v2/ >/dev/null 2>&1; then
+            echo "::error::Protected managed-image registry remained after cleanup" >&2
             exit 1
           fi
 
@@ -4571,6 +4579,42 @@ jobs:
         run: >-
           npx tsx tools/e2e/live-vitest-invocation.mts run --test-path test/e2e/live/managed-image-multiarch-startup.test.ts
 
+      - name: Validate OpenClaw managed-image security boundary
+        shell: bash
+        run: >-
+          env -u DOCKER_CONFIG -u DOCKERHUB_USERNAME -u DOCKERHUB_TOKEN npx vitest run --project integration test/e2e-runtime/managed-image-openclaw-security.test.ts --silent=false --reporter=default --reporter=test/e2e/risk-signal-reporter.ts
+
+      - name: Validate managed-image glibc probe lifecycle
+        if: ${{ !cancelled() }}
+        shell: bash
+        run: |
+          set -euo pipefail
+          export NEMOCLAW_RUN_GLIBC_PROBE_DOCKER_E2E=1
+          NEMOCLAW_TEST_IMAGE="$(jq -er '.[] | select(.agent == "openclaw") | .reference' "$NEMOCLAW_PROTECTED_MANAGED_IMAGE_CONTRACT")"
+          export NEMOCLAW_TEST_IMAGE
+          env -u DOCKER_CONFIG -u DOCKERHUB_USERNAME -u DOCKERHUB_TOKEN npx vitest run --project integration test/e2e-runtime/image-compatibility-docker-lifecycle.test.ts --silent=false --reporter=default --reporter=test/e2e/risk-signal-reporter.ts
+
+      - name: Remove protected managed-image cohort resources
+        if: always()
+        shell: bash
+        run: |
+          set -euo pipefail
+          label="io.nvidia.nemoclaw.managed-image.cohort=${NEMOCLAW_PROTECTED_MANAGED_IMAGE_COHORT}"
+          cleanup_failed=0
+          mapfile -t owned_containers < <(docker ps -aq --filter "label=$label")
+          ((${#owned_containers[@]} == 0)) || docker rm -f "${owned_containers[@]}" >/dev/null || cleanup_failed=1
+          mapfile -t owned_images < <(docker image ls -q --filter "label=$label" | sort -u)
+          ((${#owned_images[@]} == 0)) || docker image rm -f "${owned_images[@]}" >/dev/null || cleanup_failed=1
+          mapfile -t owned_volumes < <(docker volume ls -q --filter "label=$label")
+          ((${#owned_volumes[@]} == 0)) || docker volume rm -f "${owned_volumes[@]}" >/dev/null || cleanup_failed=1
+          if docker ps -aq --filter "label=$label" | grep -q . ||
+            docker image ls -q --filter "label=$label" | grep -q . ||
+            docker volume ls -q --filter "label=$label" | grep -q .; then
+            echo "::error::Protected managed-image cohort resources remained after cleanup" >&2
+            exit 1
+          fi
+          ((cleanup_failed == 0)) || exit 1
+
       - name: Publish exact amd64 protected runtime build cache
         if: ${{ matrix.platform == 'linux/amd64' }}
         uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
@@ -5086,7 +5130,7 @@ jobs:
           fi
           docker run --detach \
             --name "$NEMOCLAW_PROTECTED_REGISTRY_NAME" \
-            --label "io.nvidia.nemoclaw.e2e-owner=${NEMOCLAW_PROTECTED_MANAGED_IMAGE_COHORT}" \
+            --label "io.nvidia.nemoclaw.managed-image.cohort=${NEMOCLAW_PROTECTED_MANAGED_IMAGE_COHORT}" \
             --label "io.nvidia.nemoclaw.e2e-platform=linux/amd64" \
             --publish 127.0.0.1:5000:5000 \
             docker.io/library/registry@sha256:a3d8aaa63ed8681a604f1dea0aa03f100d5895b6a58ace528858a7b332415373
@@ -5146,7 +5190,7 @@ jobs:
           if docker container inspect "$NEMOCLAW_PROTECTED_REGISTRY_NAME" >/dev/null 2>&1; then
             owner="$(
               docker container inspect \
-                --format '{{index .Config.Labels "io.nvidia.nemoclaw.e2e-owner"}}' \
+                --format '{{index .Config.Labels "io.nvidia.nemoclaw.managed-image.cohort"}}' \
                 "$NEMOCLAW_PROTECTED_REGISTRY_NAME"
             )"
             [[ "$owner" == "$NEMOCLAW_PROTECTED_MANAGED_IMAGE_COHORT" ]] || {

.github/workflows/main.yaml

@@ -348,7 +348,7 @@ jobs:
       - real-openclaw-dist-harness
       - cli-tests
       - plugin-tests
-      - sandbox-images-and-e2e
+      - sandbox-image-contracts
     if: always()
     permissions:
       actions: read
@@ -366,7 +366,7 @@ jobs:
           CLI_TESTS_RESULT: ${{ needs['cli-tests'].result }}
           GH_TOKEN: ${{ github.token }}
           PLUGIN_TESTS_RESULT: ${{ needs['plugin-tests'].result }}
-          SANDBOX_IMAGES_E2E_RESULT: ${{ needs['sandbox-images-and-e2e'].result }}
+          SANDBOX_IMAGE_CONTRACTS_RESULT: ${{ needs['sandbox-image-contracts'].result }}
           RUN_ATTEMPT: ${{ github.run_attempt }}
           RUN_ID: ${{ github.run_id }}
           RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
@@ -404,7 +404,7 @@ jobs:
             local job_id=""
             dependency_url="$RUN_URL"
             case "$name" in
-              static-checks|build-typecheck|installer-integration|wechat-runtime-audit|reviewed-npm-audit|real-openclaw-dist-harness|cli-tests|plugin-tests|sandbox-images-and-e2e) ;;
+              static-checks|build-typecheck|installer-integration|wechat-runtime-audit|reviewed-npm-audit|real-openclaw-dist-harness|cli-tests|plugin-tests|sandbox-image-contracts) ;;
               *) return ;;
             esac
             load_job_listing
@@ -447,13 +447,13 @@ jobs:
           require_success "real-openclaw-dist-harness" "$REAL_OPENCLAW_DIST_HARNESS_RESULT"
           require_success "cli-tests" "$CLI_TESTS_RESULT"
           require_success "plugin-tests" "$PLUGIN_TESTS_RESULT"
-          require_success "sandbox-images-and-e2e" "$SANDBOX_IMAGES_E2E_RESULT"
+          require_success "sandbox-image-contracts" "$SANDBOX_IMAGE_CONTRACTS_RESULT"
 
           [ "$failed" -eq 0 ]
 
-  sandbox-images-and-e2e:
+  sandbox-image-contracts:
     needs: [static-checks, build-typecheck]
-    uses: ./.github/workflows/sandbox-images-and-e2e.yaml
+    uses: ./.github/workflows/sandbox-images.yaml
     secrets:
       DOCKERHUB_USERNAME: ${{ secrets.DOCKERHUB_USERNAME }}
       DOCKERHUB_TOKEN: ${{ secrets.DOCKERHUB_TOKEN }}

.github/workflows/pr-self-hosted.yaml

@@ -204,21 +204,8 @@ jobs:
           scripts/check-production-build-args.sh "${build_args[@]}"
           docker build "${build_args[@]}" -t nemoclaw-production .
 
-      - name: Build sandbox test image (fixtures layered on production)
-        run: docker build -f test/Dockerfile.sandbox --build-arg BASE_IMAGE=nemoclaw-production -t nemoclaw-sandbox-test .
-
-      - name: Save images to tarballs
-        run: |
-          docker save nemoclaw-sandbox-test | gzip > /tmp/sandbox-test-image.tar.gz
-          docker save nemoclaw-production | gzip > /tmp/isolation-image.tar.gz
-
-      - name: Upload sandbox test image
-        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
-        with:
-          name: sandbox-test-image
-          path: /tmp/sandbox-test-image.tar.gz
-          retention-days: 1
-          if-no-files-found: error
+      - name: Save production image
+        run: docker save nemoclaw-production | gzip > /tmp/isolation-image.tar.gz
 
       - name: Upload isolation image
         uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
@@ -249,40 +236,29 @@ jobs:
           scripts/check-production-build-args.sh "${build_args[@]}"
           docker build "${build_args[@]}" -t nemoclaw-production-arm64 .
 
-      - name: Build sandbox test image on arm64
-        run: docker build -f test/Dockerfile.sandbox --build-arg BASE_IMAGE=nemoclaw-production-arm64 -t nemoclaw-sandbox-test-arm64 .
 
-  test-e2e-sandbox:
+  managed-image-openclaw-security:
     runs-on: linux-amd64-cpu4
-    timeout-minutes: 15
+    timeout-minutes: 10
     needs: build-sandbox-images
+    env:
+      E2E_ARTIFACT_DIR: ${{ github.workspace }}/e2e-artifacts/live/managed-image-openclaw-security
+      E2E_TARGET_ID: managed-image-openclaw-security
+      NEMOCLAW_E2E_SHARD: default
+      NEMOCLAW_MANAGED_IMAGE_SECURITY_COHORT: pr-${{ github.run_id }}-${{ github.run_attempt }}
+      NEMOCLAW_RUN_LIVE_E2E: "1"
+      NEMOCLAW_TEST_IMAGE: nemoclaw-production
     steps:
       - name: Checkout
         uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
         with:
           persist-credentials: false
 
-      - name: Download image artifact
-        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
-        with:
-          name: sandbox-test-image
-          path: /tmp
-
-      - name: Load image
-        run: gunzip -c /tmp/sandbox-test-image.tar.gz | docker load
-
-      - name: Run sandbox E2E tests
-        run: docker run --rm -v "${{ github.workspace }}/test:/opt/test" nemoclaw-sandbox-test /opt/test/e2e-test.sh
-
-  test-e2e-gateway-isolation:
-    runs-on: linux-amd64-cpu4
-    timeout-minutes: 15
-    needs: build-sandbox-images
-    steps:
-      - name: Checkout
-        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
-        with:
-          persist-credentials: false
+      - name: Bind managed-image risk signal identity
+        run: |
+          set -euo pipefail
+          echo "NEMOCLAW_E2E_EXPECTED_SHA=$(git rev-parse HEAD)" >> "$GITHUB_ENV"
+          echo "NEMOCLAW_E2E_CORRELATION_ID=$(python3 -c 'import uuid; print(uuid.uuid4())')" >> "$GITHUB_ENV"
 
       - name: Set up Node
         uses: actions/setup-node@820762786026740c76f36085b0efc47a31fe5020 # v7.0.0
@@ -291,7 +267,7 @@ jobs:
           cache: npm
 
       - name: Install root dependencies
-        run: npm ci --ignore-scripts
+        run: npm ci --ignore-scripts --no-audit --no-fund
 
       - name: Download image artifact
         uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8.0.1
@@ -300,20 +276,39 @@ jobs:
           path: /tmp
 
       - name: Load image
-        run: |
-          gunzip -c /tmp/isolation-image.tar.gz | docker load
-          docker image inspect nemoclaw-production >/dev/null
+        run: gunzip -c /tmp/isolation-image.tar.gz | docker load
 
-      - name: Run glibc probe lifecycle regression
+      - name: Validate OpenClaw managed-image security boundary
+        run: >-
+          npx vitest run --project integration test/e2e-runtime/managed-image-openclaw-security.test.ts --silent=false --reporter=default --reporter=test/e2e/risk-signal-reporter.ts
+
+      - name: Validate glibc probe lifecycle
+        if: ${{ !cancelled() }}
         env:
           NEMOCLAW_RUN_GLIBC_PROBE_DOCKER_E2E: "1"
-          NEMOCLAW_TEST_IMAGE: nemoclaw-production
-        run: npx vitest run --project integration test/e2e-runtime/image-compatibility-docker-lifecycle.test.ts --silent=false --reporter=default
-
-      - name: Run gateway isolation E2E tests
-        run: NEMOCLAW_TEST_IMAGE=nemoclaw-production bash test/e2e-gateway-isolation.sh
+        run: >-
+          npx vitest run --project integration test/e2e-runtime/image-compatibility-docker-lifecycle.test.ts
+          --silent=false --reporter=default --reporter=test/e2e/risk-signal-reporter.ts
 
-  test-e2e-port-overrides:
+      - name: Remove managed-image security resources
+        if: ${{ always() }}
+        run: |
+          set -euo pipefail
+          label="io.nvidia.nemoclaw.managed-image.cohort=$NEMOCLAW_MANAGED_IMAGE_SECURITY_COHORT"
+          cleanup_failed=0
+          mapfile -t containers < <(docker ps -aq --filter "label=$label")
+          ((${#containers[@]} == 0)) || docker rm -f "${containers[@]}" >/dev/null || cleanup_failed=1
+          mapfile -t volumes < <(docker volume ls -q --filter "label=$label")
+          ((${#volumes[@]} == 0)) || docker volume rm -f "${volumes[@]}" >/dev/null || cleanup_failed=1
+          docker ps -aq --filter "label=$label" | grep -q . && cleanup_failed=1
+          docker volume ls -q --filter "label=$label" | grep -q . && cleanup_failed=1
+          exit "$cleanup_failed"
+
+      - name: Upload OpenClaw managed-image security evidence
+        if: ${{ always() }}
+        uses: ./.github/actions/upload-e2e-artifacts
+
+  port-override-image-contract:
     runs-on: linux-amd64-cpu4
     timeout-minutes: 10
     needs: build-sandbox-images