Overview

Product video
Most enterprises can build an AI agent. Far fewer can put one into production, because the controls their security, risk and compliance functions require do not exist between the agent and the company's data. Xnode Cortx closes that gap. It is an AI Control Plane that supplies identity, access control, guardrails and audit as infrastructure, so an agent inherits the same permissions and accountability as the person it acts for.
GOVERNANCE ACROSS FOUR AXES
Identity and access. Single sign on through OIDC, SAML, Microsoft Entra, Active Directory and LDAP, with SCIM lifecycle management and support for multiple identity providers. The end user's own identity is carried through to every downstream system, so the source system's permissions still apply instead of being bypassed by a shared service account.
Policy and guardrails. A hierarchical policy model spanning organisation, department, workspace, model group, team and individual user. Detection categories with configurable pattern libraries, maker checker approval on policy changes, and a resolver view that shows exactly which policy determined each outcome for a named user, with drill down to the rule that fired.
Model eligibility. Data classification is applied to the data object rather than only the request, so content that must never reach an external model is blocked from doing so regardless of who asks or which agent is running.
Tool and data access. An MCP Gateway that governs which tools an agent can reach and under whose credentials, with access control extended across databases, SaaS applications, file stores and retrieval pipelines. Read only and full access toolkits can be assigned separately.
OBSERVABILITY AND CONTROL
A tamper evident, hash chained record of every allow and deny. Agent and model level cost tracking with budget enforcement. Agent evaluation against the customer's own data sets, comparing models side by side on the same questions. Discovery of unsanctioned AI use across the environment.
DEPLOYMENT
Xnode Cortx runs entirely within the customer's environment on Kubernetes, on premises or in the customer's own cloud tenant. Supported patterns are controlled egress, sovereign cloud and fully air gapped. No customer data leaves the customer's perimeter, and no ongoing connection to XNODE is required to operate the platform.
COMPLIANCE
SOC 2 Type II, ISO/IEC 27001, ISO/IEC 42001, GDPR and HIPAA.
WHO IT IS FOR
Financial services, government and public sector, healthcare and manufacturing. Typical buyers are CISOs, heads of data, and AI platform teams who need agentic AI in production under the same controls that already govern their other systems.
Highlights
- Highlight 1 Governed agentic AI without your data leaving your network. Xnode Cortx deploys on premises, in sovereign cloud, or fully air gapped, on your own Kubernetes. No ongoing connection to XNODE is required to operate the platform.
- Highlight 2 Every agent action carries a real identity. Per user identity is passed through to each downstream system so existing permissions still apply, policy is enforced at the tool and retrieval boundary, and every allow and deny is written to a tamper evident, hash chained audit record.
- Highlight 3 Control which data may reach which model. Classification is applied to the data object itself, so restricted content is blocked from external models regardless of who asks or which agent is running, with a resolver view showing exactly which policy decided each outcome.
Details
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
Pricing
Vendor refund policy
This is a Bring Your Own License product. AWS Marketplace does not charge for use of the software, so no refunds are processed through AWS Marketplace. Licensing terms, including any refund provisions, are governed by the agreement between the customer and XNODE INC. For licensing questions, contact support@xnode.ai .
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
Xnode Cortx Platform — Infrastructure Layer
- Amazon EKS Anywhere
Helm chart
Helm charts are Kubernetes YAML manifests combined into a single package that can be installed on Kubernetes clusters. The containerized application is deployed on a cluster by running a single Helm install command to install the seller-provided Helm chart.
Version release notes
Xnode Cortx v1.1.0 - AWS Marketplace Release
Xnode Cortx is an AI Control Plane for governed agentic AI. This is the first release of Xnode Cortx on AWS Marketplace, delivering the full v1.1.0 platform including all infrastructure services and AI application microservices.
WHAT'S INCLUDED
Platform Infrastructure (cortx-platform): PostgreSQL, Redis, RabbitMQ, MinIO, Neo4j, HashiCorp Vault, NATS, Qdrant, Temporal, Nexus, OpenFGA, Phoenix, and full observability stack (Prometheus, Grafana, Loki, Mimir, Tempo, Pyroscope, OTel Collector). Automated Vault bootstrap, database initialisation, and post-install configuration on first install.
AI Application Services (cortx-apps): 31 microservices including agentic APIs, guardrails, RBAC, identity, integrations, web UI, and coding assistant. Requires a valid CORTX license bundle - contact support@xnode.ai .
GUARDRAILS AND POLICY ENFORCEMENT
- Model-narrowed Global policies enforce only for the selected models or model groups.
- Mapped pattern navigation in Protection Categories routes correctly to the Pattern Library.
CONVERSATIONS AND FILE UPLOADS
- All files uploaded in a conversation are considered when responding, not only the most recently uploaded file.
JIRA INTEGRATION
- Jira credentials are sent in the request body, not the URL.
- On-premises Jira operations route to the correct REST API (v2) with real error diagnostics.
AIR GAPPED AND ON PREMISES CONNECTIVITY
- CORTX_AIRGAP_TLS_VERIFY setting controls TLS verification for internal service-to-service calls behind a private CA.
PLATFORM RELIABILITY
- Database connection leak in Teams middleware resolved.
REQUIREMENTS
- Kubernetes 1.27+ (EKS recommended)
- Minimum 8 vCPU / 32 GB RAM
- EFS storage class (EKS)
Support: support@xnode.ai | https://xnode.ai/contact/ © 2026 Xnode Inc. All rights reserved.
Additional details
Usage instructions
PREREQUISITES
Kubernetes 1.27+ (EKS recommended) Helm 3.12+, kubectl, AWS CLI Minimum 8 vCPU / 32 GB RAM across worker nodes EFS storage class configured (EKS) or equivalent RWX storage class STEP 1 - Authenticate to ECR aws ecr get-login-password --region us-east-1 | helm registry login --username AWS --password-stdin 709825985650.dkr.ecr.us-east-1.amazonaws.com
STEP 2 - Pull default values helm show values oci://709825985650.dkr.ecr.us-east-1.amazonaws.com/xnode/cortx-platform --version 1.1.0-rc1 > my-platform-values.yaml
STEP 3 - Edit my-platform-values.yaml Fill all fields:
global: envName: "prod" # short label, no spaces e.g. prod, staging domain: "yourdomain.com" # base domain for ingress hostnames clusterType: "eks" # eks | self-managed storageClass: "efs-sc" # EKS: efs-sc | self-managed: nfs-storage ingressClass: "alb" # EKS: alb | self-managed: nginx
secrets: passwords: dbAdmin: "<min 12 chars, no # character>" dbEditor: "<min 12 chars>" dbReader: "<min 12 chars>" minioSecret: "<min 12 chars>" rabbitmq: "<min 12 chars, no # character>" redis: "<min 12 chars>" neo4j: "<min 12 chars>" grafana: "<min 12 chars>" phoenix: "<min 12 chars>" apiKeys: internal: "<run: openssl rand -hex 32>" btob: "<run: uuidgen>" ingestion: "<run: openssl rand -hex 32>" phoenix: "<run: openssl rand -hex 32>" qdrant: "<run: openssl rand -hex 32>" llm: openaiKey: "" smtp: host: "" port: "25" from: "noreply@yourdomain.com "
onboarding: adminEmail: "admin@yourdomain.com " entityName: "Your Organisation" accountName: "Your Account" firstName: "Admin" lastName: "User" mobileNumber: "1234567890" country: "US" countryCode: "US"
STEP 4 - Install helm install cortx-platform oci://709825985650.dkr.ecr.us-east-1.amazonaws.com/xnode/cortx-platform --version 1.1.0-rc1 --namespace cortx-platform --create-namespace -f my-platform-values.yaml
STEP 5 - Wait for all pods to be Running kubectl get pods -n cortx-platform -w kubectl get pods -n vault -w kubectl get pods -n postgres -w
All pods must show Running or Completed before installing cortx-apps.
STEP 6 - Retrieve Vault token (needed for cortx-apps) kubectl get secret vault-unseal-keys -n vault -o jsonpath='{.data.root-token}' | base64 -d
Save this token - you will need it as vaultToken in cortx-apps values.
TROUBLESHOOTING
ImagePullBackOff: re-run Step 1 to re-authenticate to ECR Pod CrashLoopBackOff: kubectl describe pod -n cortx-platform Vault not initialising: kubectl logs -n vault vault-0 Support: support@xnode.ai
Resources
Vendor resources
Support
Vendor support
XNODE provides direct support for Xnode Cortx from the engineering team that builds it.
Email: support@xnode.ai Web: https://xnode.ai/contact/
Standard support is available Monday to Friday during business hours, with a 30 minute initial response target for critical severity issues. Enhanced support terms, including extended coverage hours and named technical contacts, are available under a private offer.
Deployment assistance, environment sizing guidance and installation support are included with every deployment. Because Xnode Cortx runs inside the customer's own network, XNODE has no access to customer environments or data, and all support is delivered with the customer's own team in control of their environment.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.