atlas-iac/services/hermes/deployment.yaml

# services/hermes/deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: hermes
  namespace: hermes
  labels:
    app: hermes
spec:
  replicas: 1
  revisionHistoryLimit: 2
  progressDeadlineSeconds: 2700
  strategy:
    type: Recreate
  selector:
    matchLabels:
      app: hermes
  template:
    metadata:
      labels:
        app: hermes
      annotations:
        ai.bstein.dev/frontend-fix: scope PTY attachment by selected conversation
        ai.bstein.dev/router-wire-contract: ollama-numeric-keepalive
        ai.bstein.dev/model: anthropic/claude-opus-5, falling back to openai-codex/gpt-5.6-terra then titan-20 Qwen 14B
        ai.bstein.dev/role: testing-triage
        ai.bstein.dev/placement: titan-21 preferred, Jetson preferred, arm64 fallback
        ai.bstein.dev/config-rev: "20260815-runtime-access-boundary"
        vault.hashicorp.com/agent-inject: "true"
        vault.hashicorp.com/role: hermes
        vault.hashicorp.com/agent-inject-secret-triage-api-key: kv/data/atlas/hermes/triage-api
        vault.hashicorp.com/agent-inject-template-triage-api-key: |
          {{- with secret "kv/data/atlas/hermes/triage-api" -}}
          {{ .Data.data.api_key }}
          {{- end }}
        vault.hashicorp.com/agent-pre-populate-only: "true"
        # Without this the agent init container is appended after the
        # pod's own init containers, so init-config runs before the
        # secret file exists and silently falls back to the Secret -
        # the migration looks complete while Vault is never read.
        vault.hashicorp.com/agent-init-first: "true"
        vault.hashicorp.com/agent-requests-cpu: 25m
        vault.hashicorp.com/agent-requests-mem: 32Mi
        vault.hashicorp.com/agent-limits-cpu: 100m
        vault.hashicorp.com/agent-limits-mem: 128Mi
    spec:
      serviceAccountName: hermes-triage
      automountServiceAccountToken: true
      securityContext:
        fsGroup: 10000
        seccompProfile:
          type: RuntimeDefault
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
              - matchExpressions:
                  - key: kubernetes.io/arch
                    operator: In
                    values:
                      - arm64
                  - key: node-role.kubernetes.io/worker
                    operator: In
                    values:
                      - "true"
                  - key: kubernetes.io/hostname
                    operator: NotIn
                    values:
                      - titan-08
                      - titan-13
                      - titan-14
                      - titan-17
                      - titan-18
          preferredDuringSchedulingIgnoredDuringExecution:
            - weight: 80
              preference:
                matchExpressions:
                  - key: atlas.bstein.dev/spillover
                    operator: DoesNotExist
            - weight: 60
              preference:
                matchExpressions:
                  - key: hardware
                    operator: In
                    values:
                      - rpi5
            - weight: 50
              preference:
                matchExpressions:
                  - key: hardware
                    operator: In
                    values:
                      - rpi4
      initContainers:
        - name: init-config
          image: busybox:1.37
          imagePullPolicy: IfNotPresent
          command:
            - sh
            - -c
            - |
              set -eu
              mkdir -p /opt/data/workspace/triage-proof /opt/data/home/.local/bin /opt/data/logs
              cp /config/config.yaml /opt/data/config.yaml
              cp /config/SOUL.md /opt/data/SOUL.md
              cp /config/AGENTS.md /opt/data/workspace/AGENTS.md
              cp /config/START-HERE.md /opt/data/workspace/START-HERE.md
              cp /config/HERMES-CAPABILITIES.md /opt/data/workspace/HERMES-CAPABILITIES.md
              cp /guide/OPERATOR-RUNBOOK.md /opt/data/workspace/HERMES-OPERATOR-RUNBOOK.md
              cp /config/ATLAS-TRIAGE-PROOFS.md /opt/data/workspace/triage-proof/ATLAS-TRIAGE-PROOFS.md
              cp /config/START-HERE.md /opt/data/START-HERE.md
              cp /config/HERMES-CAPABILITIES.md /opt/data/HERMES-CAPABILITIES.md
              cp /guide/OPERATOR-RUNBOOK.md /opt/data/HERMES-OPERATOR-RUNBOOK.md
              cp /config/ATLAS-TRIAGE-PROOFS.md /opt/data/ATLAS-TRIAGE-PROOFS.md
              touch /opt/data/.env
              for key in ANTHROPIC_API_KEY API_SERVER_KEY CLAUDE_API_KEY CLAUDE_CODE_OAUTH_TOKEN GITEA_TOKEN GITEA_USERNAME HERMES_IMAGE_BROKER_KEY OPENAI_API_KEY; do
                grep -v "^${key}=" /opt/data/.env > /opt/data/.env.tmp || true
                mv /opt/data/.env.tmp /opt/data/.env
              done
              rm -f /opt/data/auth.json
              chmod 0600 /opt/data/.env
              chown -R 10000:10000 /opt/data
          securityContext:
            runAsUser: 0
            runAsGroup: 0
          volumeMounts:
            - name: home
              mountPath: /opt/data
            - name: config
              mountPath: /config
            - name: operator-guide
              mountPath: /guide
          resources:
            requests:
              cpu: 25m
              memory: 32Mi
            limits:
              cpu: 100m
              memory: 64Mi
        - name: stage-runtime-access
          image: registry.bstein.dev/bstein/hermes-agent@sha256:81970563e542f0720773e72297810b3a844b83e381e278f25c0916c78d930107
          imagePullPolicy: IfNotPresent
          command:
            - /opt/hermes/.venv/bin/python
            - /opt/coordinator/stage_runtime_access.py
            - triage
          securityContext:
            allowPrivilegeEscalation: false
            runAsUser: 0
            runAsGroup: 0
            seccompProfile:
              type: RuntimeDefault
          volumeMounts:
            - {name: coordinator, mountPath: /opt/coordinator, readOnly: true}
            - {name: runtime-access, mountPath: /runtime-access}
          resources:
            requests: {cpu: 25m, memory: 32Mi}
            limits: {cpu: 100m, memory: 64Mi}
        - name: patch-auth
          image: registry.bstein.dev/bstein/hermes-agent@sha256:81970563e542f0720773e72297810b3a844b83e381e278f25c0916c78d930107
          imagePullPolicy: IfNotPresent
          command:
            - /opt/hermes/.venv/bin/python
            - /opt/coordinator/patch_hermes_auth.py
            - /opt/hermes/hermes_cli/auth.py
            - /patched/auth.py
          securityContext:
            allowPrivilegeEscalation: false
            runAsUser: 10000
            runAsGroup: 10000
            seccompProfile:
              type: RuntimeDefault
          volumeMounts:
            - name: coordinator
              mountPath: /opt/coordinator
              readOnly: true
            - name: auth-patch
              mountPath: /patched
          resources:
            requests:
              cpu: 25m
              memory: 64Mi
            limits:
              cpu: 100m
              memory: 128Mi
        - name: patch-api-server-sessions
          image: registry.bstein.dev/bstein/hermes-agent@sha256:81970563e542f0720773e72297810b3a844b83e381e278f25c0916c78d930107
          imagePullPolicy: IfNotPresent
          command:
            - /bin/sh
            - -ec
            - |
              /opt/hermes/.venv/bin/python /opt/coordinator/patch_api_server_sessions.py \
                /opt/hermes/gateway/platforms/api_server.py /patched/api_server.py
              /opt/hermes/.venv/bin/python /opt/coordinator/migrate_api_session_lineage.py
          env:
            - {name: HERMES_API_DEFAULT_PARENT_SESSION_ID, value: automated-triage}
            - name: HERMES_API_DEFAULT_PARENT_MATCH_PREFIXES
              value: "A static-analysis finding, not a build failure.||Use $triage-titan-test-failures."
          securityContext:
            allowPrivilegeEscalation: false
            runAsUser: 10000
            runAsGroup: 10000
            seccompProfile:
              type: RuntimeDefault
          volumeMounts:
            - {name: home, mountPath: /opt/data}
            - {name: coordinator, mountPath: /opt/coordinator, readOnly: true}
            - {name: api-server-patch, mountPath: /patched}
          resources:
            requests: {cpu: 25m, memory: 64Mi}
            limits: {cpu: 100m, memory: 128Mi}
        - name: patch-subprocess-secret-boundary
          image: registry.bstein.dev/bstein/hermes-agent@sha256:81970563e542f0720773e72297810b3a844b83e381e278f25c0916c78d930107
          imagePullPolicy: IfNotPresent
          command:
            - /bin/sh
            - -ec
            - |
              /opt/hermes/.venv/bin/python /opt/coordinator/patch_subprocess_secret_boundary.py \
                /opt/hermes/tools/environments/local.py /patched/local.py
              /opt/hermes/.venv/bin/python /opt/coordinator/patch_process_output_redaction.py \
                /opt/hermes/tools/process_registry.py /patched/process_registry.py
          securityContext:
            allowPrivilegeEscalation: false
            runAsUser: 10000
            runAsGroup: 10000
            seccompProfile:
              type: RuntimeDefault
          volumeMounts:
            - {name: coordinator, mountPath: /opt/coordinator, readOnly: true}
            - {name: subprocess-secret-patch, mountPath: /patched}
          resources:
            requests: {cpu: 25m, memory: 64Mi}
            limits: {cpu: 100m, memory: 128Mi}
        - name: install-kubectl
          image: bitnami/kubectl@sha256:554ab88b1858e8424c55de37ad417b16f2a0e65d1607aa0f3fe3ce9b9f10b131
          imagePullPolicy: IfNotPresent
          command:
            - /bin/sh
            - -c
            - |
              set -e
              cp "$(command -v kubectl)" /tools/kubectl
              chmod 0755 /tools/kubectl
              chown 10000:10000 /tools/kubectl
          securityContext:
            runAsUser: 0
            runAsGroup: 0
          volumeMounts:
            - name: tools
              mountPath: /tools
          resources:
            requests:
              cpu: 25m
              memory: 32Mi
            limits:
              cpu: 100m
              memory: 64Mi
      containers:
        - name: hermes
          image: registry.bstein.dev/bstein/hermes-agent@sha256:81970563e542f0720773e72297810b3a844b83e381e278f25c0916c78d930107
          imagePullPolicy: IfNotPresent
          command: [/bin/sh, -ec]
          args:
            - |
              API_SERVER_KEY="$(tr -d '\r\n' < /runtime-access/triage-api-key)"
              test -n "${API_SERVER_KEY}"
              export API_SERVER_KEY
              exec /opt/hermes/.venv/bin/hermes gateway run --no-supervise
          ports:
            - name: api
              containerPort: 8642
              protocol: TCP
          env:
            - name: HERMES_HOME
              value: /opt/data
            - name: HERMES_AUTH_FILE
              value: /runtime-access/hermes-auth.json
            - name: HOME
              value: /opt/data/home
            - name: PATH
              value: /opt/data/home/.local/bin:/opt/hermes/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
            - name: HERMES_DASHBOARD
              value: "0"
            - name: HERMES_DASHBOARD_PUBLIC_URL
              value: https://triage.bstein.dev
            - name: API_SERVER_ENABLED
              value: "true"
            - name: API_SERVER_HOST
              value: 0.0.0.0
            - name: API_SERVER_PORT
              value: "8642"
            - name: API_SERVER_CORS_ORIGINS
              value: https://triage.bstein.dev
            - name: VICTORIA_METRICS_URL
              value: http://victoria-metrics-single-server.monitoring.svc.cluster.local:8428
            - name: ARIADNE_BASE_URL
              value: http://ariadne.maintenance.svc.cluster.local
            - name: JENKINS_BASE_URL
              value: http://jenkins.jenkins.svc.cluster.local:8080
            - name: GITEA_BASE_URL
              value: https://scm.bstein.dev
            - name: GRAFANA_BASE_URL
              value: https://metrics.bstein.dev
            - name: HERMES_AUTO_ROUTER_PROFILE
              value: triage
            - name: HERMES_API_DEFAULT_PARENT_SESSION_ID
              value: automated-triage
            - name: HERMES_API_DEFAULT_PARENT_MATCH_PREFIXES
              value: "A static-analysis finding, not a build failure.||Use $triage-titan-test-failures."
          securityContext:
            allowPrivilegeEscalation: false
            runAsNonRoot: true
            runAsUser: 10000
            runAsGroup: 10000
            capabilities:
              drop: [ALL]
            seccompProfile:
              type: RuntimeDefault
          volumeMounts:
            - name: home
              mountPath: /opt/data
            - name: runtime-access
              mountPath: /runtime-access
            - name: auth-patch
              mountPath: /opt/hermes/hermes_cli/auth.py
              subPath: auth.py
            - name: api-server-patch
              mountPath: /opt/hermes/gateway/platforms/api_server.py
              subPath: api_server.py
            - name: subprocess-secret-patch
              mountPath: /opt/hermes/tools/environments/local.py
              subPath: local.py
            - name: subprocess-secret-patch
              mountPath: /opt/hermes/tools/process_registry.py
              subPath: process_registry.py
            - name: tools
              mountPath: /usr/local/bin/kubectl
              subPath: kubectl
            - name: triage-skill
              mountPath: /opt/data/workspace/skills/triage-titan-test-failures
              readOnly: true
            - name: mastery-skill
              mountPath: /opt/data/workspace/skills/master-hermes-on-atlas
              readOnly: true
            - name: service-health-skill
              mountPath: /opt/data/workspace/skills/triage-atlas-service-health
              readOnly: true
            - name: alert-tuning-skill
              mountPath: /opt/data/workspace/skills/tune-atlas-alerts
              readOnly: true
            - name: auto-router-plugin
              mountPath: /opt/data/plugins/auto-router
              readOnly: true
          startupProbe:
            tcpSocket:
              port: api
            periodSeconds: 10
            timeoutSeconds: 5
            failureThreshold: 60
          readinessProbe:
            tcpSocket:
              port: api
            initialDelaySeconds: 30
            periodSeconds: 10
            timeoutSeconds: 5
          livenessProbe:
            tcpSocket:
              port: api
            initialDelaySeconds: 90
            periodSeconds: 30
            timeoutSeconds: 10
          resources:
            requests:
              cpu: 500m
              memory: 1Gi
            limits:
              cpu: "2"
              memory: 4Gi
        - name: webui
          image: registry.bstein.dev/bstein/hermes-webui@sha256:ac6ba7bfd8a86227f31a9a96ebea41227ccf70391f7dcf34206f4d58e835e50e
          imagePullPolicy: IfNotPresent
          command: [/bin/sh, -ec]
          args:
            - |
              api_key="$(tr -d '\r\n' < /runtime-access/triage-api-key)"
              test -n "${api_key}"
              export API_SERVER_KEY="${api_key}"
              export HERMES_WEBUI_GATEWAY_API_KEY="${api_key}"
              exec /opt/hermes/.venv/bin/python /opt/hermes-webui/server.py
          ports:
            - {name: dashboard, containerPort: 8787, protocol: TCP}
          env:
            - {name: HERMES_HOME, value: /opt/data}
            - {name: HERMES_AUTH_FILE, value: /runtime-access/hermes-auth.json}
            - {name: HOME, value: /opt/data/home}
            - {name: HERMES_WEBUI_AGENT_DIR, value: /opt/hermes}
            - {name: HERMES_WEBUI_HOST, value: 0.0.0.0}
            - {name: HERMES_WEBUI_PORT, value: "8787"}
            - {name: HERMES_WEBUI_STATE_DIR, value: /opt/data/webui}
            - {name: HERMES_WEBUI_DEFAULT_WORKSPACE, value: /opt/data/workspace}
            - {name: HERMES_WEBUI_CHAT_BACKEND, value: gateway}
            - {name: HERMES_WEBUI_GATEWAY_BASE_URL, value: http://127.0.0.1:8642}
            - {name: HERMES_WEBUI_GATEWAY_USE_RUNS_API, value: "true"}
            - {name: HERMES_WEBUI_SKIP_ONBOARDING, value: "1"}
            - {name: HERMES_WEBUI_SECURE, value: "1"}
            - {name: HERMES_WEBUI_ALLOWED_ORIGINS, value: https://triage.bstein.dev}
            - {name: HERMES_WEBUI_TRUST_FORWARDED_HOST, value: "1"}
            - {name: HERMES_WEBUI_TRUST_FORWARDED_PROTO, value: "1"}
            - {name: HERMES_ROUTER_PROFILE, value: triage}
          volumeMounts:
            - {name: home, mountPath: /opt/data}
            - {name: runtime-access, mountPath: /runtime-access, readOnly: true}
            - {name: tmp, mountPath: /tmp}
          readinessProbe:
            httpGet: {path: /health, port: dashboard}
            initialDelaySeconds: 10
            periodSeconds: 10
            timeoutSeconds: 5
          livenessProbe:
            httpGet: {path: /health, port: dashboard}
            initialDelaySeconds: 30
            periodSeconds: 30
            timeoutSeconds: 10
          securityContext:
            allowPrivilegeEscalation: false
            capabilities:
              drop: [ALL]
            readOnlyRootFilesystem: true
            runAsNonRoot: true
            runAsUser: 10000
            runAsGroup: 10000
            seccompProfile:
              type: RuntimeDefault
          resources:
            requests: {cpu: 50m, memory: 128Mi}
            limits: {cpu: 750m, memory: 1Gi}
      volumes:
        - name: home
          persistentVolumeClaim:
            claimName: hermes-home
        - name: runtime-access
          emptyDir:
            medium: Memory
            sizeLimit: 2Mi
        - name: config
          configMap:
            name: hermes-config
        - name: operator-guide
          configMap:
            name: hermes-operator-guide
        - name: tools
          emptyDir: {}
        - name: coordinator
          configMap:
            name: hermes-coordinator
            defaultMode: 0555
        - name: auto-router-plugin
          configMap:
            name: hermes-auto-router-plugin
        - name: auth-patch
          emptyDir: {}
        - name: api-server-patch
          emptyDir: {}
        - name: subprocess-secret-patch
          emptyDir: {}
        - name: tmp
          emptyDir:
            sizeLimit: 256Mi
        - name: triage-skill
          configMap:
            name: hermes-triage-skill
            items:
              - key: SKILL.md
                path: SKILL.md
              - key: openai.yaml
                path: agents/openai.yaml
        - name: mastery-skill
          configMap:
            name: hermes-mastery-skill
            items:
              - key: SKILL.md
                path: SKILL.md
              - key: openai.yaml
                path: agents/openai.yaml
              - key: architecture.md
                path: references/architecture.md
              - key: curriculum.md
                path: references/curriculum.md
              - key: incident-drills.md
                path: references/incident-drills.md
              - key: mastery-rubric.md
                path: references/mastery-rubric.md
              - key: two-hour-proof-sprint.md
                path: references/two-hour-proof-sprint.md
        - name: service-health-skill
          configMap:
            name: hermes-service-health-skill
            items:
              - key: SKILL.md
                path: SKILL.md
              - key: openai.yaml
                path: agents/openai.yaml
              - key: service-map.md
                path: references/service-map.md
        - name: alert-tuning-skill
          configMap:
            name: hermes-alert-tuning-skill
            items:
              - key: SKILL.md
                path: SKILL.md
              - key: openai.yaml
                path: agents/openai.yaml
              - key: alert-review.md
                path: references/alert-review.md