The Containment Era is here. →Explore

Live · Jun 24, 2026
AWS BlueprintOpenClaw · NemoClaw · Hermes (any harness)

Agent harnesses are privileged non-human operators. Contain them.

OpenClaw, NemoClaw, and Hermes can read sensitive data, call APIs, execute code, browse the web, and persist state across sessions — without a human prompting every step. This VCA places the agent runtime in a private AWS subnet behind Aviatrix, enforces a default-deny egress policy with Distributed Cloud Firewall, and logs every connection in CoPilot FlowIQ with agent identity. No HTTPS proxy. No SDK change. No redeploy.

AWS blueprint · Default-deny egressVM shape · K8s extensionNo HTTPS proxy · No agent changesMonitor-first rollout
The threat this solves

The problem isn’t hallucination. It’s unbounded network reachability.

Agent harnesses are non-human operators that decide which API to call, which file to read, which command to execute, and which endpoint to contact — without prompting every step. A prompt injection, malicious skill, compromised dependency, or unsafe browser page can redirect that authority toward exfiltration or lateral movement. In-harness allowlists sit inside the runtime an attacker may already control. Aviatrix enforcement sits outside it.

01

Prompt injection → exfiltration

Works with prompt guardrails

Untrusted content in a web page, ticket, document, or email redirects the agent to send retrieved data to an attacker-controlled host. Indistinguishable from legitimate agent activity at the process layer.

02

Supply-chain / skill compromise

Works with dependency scanning

A malicious package, plugin, or skill phones home or downloads a payload from inside the terminal environment. No CVE at install time — package scanners pass it. The agent runtime now executes attacker-controlled code.

03

Shadow model call

Works with IAM / model gateway

Agent code or manipulated instructions send prompts or retrieved data to an unapproved model provider — OpenAI, Anthropic, Mistral, DeepSeek, or others. The named deny rule at priority 10 fires before any allow rule.

04

DNS exfiltration

Works with EDR / DNS tools

Encoded data leaves through an external DNS resolver on UDP/TCP 53 — bypassing controls that only inspect HTTPS. Rules 20/21 deny external DNS to any destination other than the VPC resolver.

05

Lateral Movement

✓ Blocked

A compromised agent tries to reach your databases or other internal servers. It's blocked from moving sideways into the rest of your network.

06

Exfiltration attempt — blocked

✓ Aviatrix DCF default-deny blocks

The agent initiates outbound HTTPS to an attacker-controlled endpoint. TCP SYN traverses the Aviatrix Spoke Gateway. Destination is not in any permit WebGroup — POST_RULES default-deny fires. Connection never completes.

07

Audit trail

✓ CoPilot FlowIQ · human-readable rule name

Blocked connection logged at the Spoke Gateway with agent subnet identity, destination FQDN or IP, policy rule name, action, and timestamp. CoPilot FlowIQ provides per-agent class traffic analysis with full attribution for SOC 2, HIPAA, PCI-DSS, FedRAMP, and DORA audit requirements.

How it works

One enforcement point. Outside the harness trust boundary.

Enforcement runs at the cloud network layer, outside the runtime the agent can influence. No HTTPS proxy to configure, no SDK to rewrite, no agent-code change required.

Layer 01 — Private subnet

Transparent egress enforcement — zero bypass risk

The agent runtime (OpenClaw / NemoClaw / Hermes) runs in a private AWS subnet with no public IP and no internet gateway route. The subnet default route 0.0.0.0/0 points to the Aviatrix Spoke Gateway — every outbound connection is intercepted transparently. Administration is through SSM Session Manager; no inbound rule and no public SSH required. No software process inside the agent can circumvent the route table.

Layer 02 — Distributed Cloud Firewall

Ordered policy evaluation · shadow-model deny at priority 10

Traffic passing through the Spoke Gateway is evaluated against an ordered DCF policy list. SmartGroups identify the source (agent subnet CIDR). WebGroups define approved destinations per agent class. The shadow-model deny rule fires at priority 10, ahead of all permit rules. The POST_RULES default-deny catches every destination not explicitly declared. In monitor mode, every rule logs but permits — baseline real agent workflows first, then enforce.

Layer 03 — Kubernetes extension

Same policy model · pod identity via SmartGroups

The same SmartGroup / WebGroup model extends to EKS. Map agent pods to namespace or label-based SmartGroups; policy follows pod identity instead of static pod IPs. Requires Controller 9.0+ for TLS decryption, URL-path filtering, and egress IDS/IPS on the K8s shape.

Enforcement architecture — AWS VM shape

Private subnet · Aviatrix Spoke Gateway · DCF policy enforcement · Zero bypass risk
PRIVATE SUBNET · AWS VPC · NO PUBLIC IP · SSM-ONLY MANAGEMENTCoding agentOpenClaw / NemoClawResearch agentNemoClaw / HermesSupport / regulatedany harness0.0.0.0/0route tableAviatrix Spoke GatewayDCF · SmartGroups · WebGroupsPriority 10: shadow-model denyPOST_RULES: default-denyZero bypass riskIn permitlist?YESPermitted egress — loggedModel gateway (NVIDIA / enterprise)OpenClaw core · SaaS APIs · MCPPackage registries (coding class only)AWS infra endpoints (SSM, ECR, STS, Logs)NODENY + LOGShadow models · C2 · external DNSUnapproved SaaS · unlisted destinationsPriority 10 names shadow-model attemptsCoPilot FlowIQAgent subnet identity · rule nameDestination · action · timestampAll allowed + denied flowsaudittrail
Permitted egressDenied + loggedRoute table enforces 0.0.0.0/0 → Spoke Gateway · no software bypass possible
Policy pack

Ordered evaluation. Shadow-model deny at priority 10.

Rules evaluate top to bottom. The vca-openclaw-shadow-model-deny rule fires at priority 10, ahead of all permit rules. WebGroups are SNI/FQDN-oriented — exact domain names, not wildcards, ports, or URL paths. In monitor mode every rule logs but permits.

Rules (in priority order)

PriorityRule nameActionDestinationNotes
10vca-openclaw-shadow-model-denyDENY + LOGwg-unapproved-model-providersapi.openai.com, api.anthropic.com, generativelanguage.googleapis.com, api.mistral.ai, openrouter.ai, api.deepseek.com … Fired before any allow rule.
18/19allow-vpc-dns-udp/tcpPERMITsg-vpc-dns-resolver UDP/TCP 53VPC resolver only. Precedes external DNS deny.
20/21deny-dns-exfil-udp/tcpDENY + LOGAny UDP/TCP 53 except VPC resolverCloses DNS tunneled exfiltration channel.
30allow-aws-infraPERMITSSM, EC2, Logs, STS, ECR/S3 endpointsRegion-generated from aws_region variable.
40/41allow-model-gatewaysPERMITsg-approved-model-gateways TCP 443integrate.api.nvidia.com, inference-api.nvidia.com, or enterprise gateway.
50allow-corePERMITwg-openclaw-core TCP 443openclaw.ai, docs.openclaw.ai, clawhub.ai, www.nvidia.com
60allow-packagesPERMITwg-approved-package-registries TCP 443registry.npmjs.org, pypi.org, github.com, *.githubusercontent.com, huggingface.co, registry-1.docker.io — coding-class agents only.
70/80allow-saas / allow-mcpPERMITwg-approved-saas-apis & wg-approved-mcp-gatewaysCRM, ticketing, approved KB, and MCP gateways — defined per agent class.
100deny-eastwestDENY + LOGAdjacent / internal CIDRsIsolates agent VPC from adjacent workloads. Extendable with SmartGroup east-west microsegmentation.
POSTDefault actionDENY + LOGAny unmatched destinationAll protocols. Every match logged to CoPilot FlowIQ with identity, destination, rule name, and timestamp.

SmartGroup and WebGroup objects

ObjectTypePurpose
sg-agent-vm-subnetCIDR SmartGroupSource identity for the VM or container shape. Scoped to the private agent subnet CIDR.
sg-vpc-dns-resolverCIDR SmartGroupVPC DNS resolver destination. Permitted before external DNS is denied.
sg-approved-model-gatewaysCIDR / FQDNEnterprise or NVIDIA model gateway. Prefer an enterprise gateway where available.
wg-openclaw-coreWebGroupOpenClaw / NemoClaw core services, docs, and required terminal UI endpoints.
wg-approved-package-registriesWebGroupnpm, PyPI, GitHub, Hugging Face, Docker Hub — coding-class agents only.
wg-approved-saas-apisWebGroupCRM, ticketing, document stores, and workflow APIs defined per agent class.
wg-approved-mcp-gatewaysWebGroupEnterprise MCP / tool gateways without flat internal network reachability.
wg-unapproved-model-providersWebGroupKnown unapproved model APIs. Denied at priority 10, ahead of all allow rules.

Agent-class presets

Agent classTypical permitsTypical denies
Locked-down demoModel gateway, OpenClaw/NemoClaw core, AWS private opsPackage registries, broad web, SaaS APIs
Coding agentGitHub, package registries, internal artifacts, model gateway, docsShadow models, external DNS, unapproved SaaS, production networks
Research agentApproved search/data APIs, document stores, model gateway, telemetryPackage registries unless needed, arbitrary uploads
Support agentCRM, ticketing, approved KB, MCP gateway, model gatewaySource-code systems and package registries unless required
Regulated-data agentSpecific internal APIs, approved model gateway, observabilityPublic package registries, public SaaS, external upload, shadow models
Assets & documentation

Everything your team needs.

Requires Aviatrix Enterprise · Controller 8.1+ · 1 managed network per deploymentNew to Aviatrix? Start the Enterprise free trial — VCAs included at no extra cost. Already deployed? Pull the Terraform from GitHub.
Technical Brief

AWS Blueprint Reference

Prerequisites, VPC and subnet design, Aviatrix Spoke Gateway insertion, full DCF ruleset in priority order, SmartGroup and WebGroup definitions, agent-class presets, validation tests, and monitor-to-enforce rollout sequence. For platform engineers deploying the VM or K8s shape.

Download PDF →
Security Architecture Brief

Threat Model & Enforcement

Agent harness threat model, full kill chain with point of intervention, control boundary map showing what enforces where, why in-harness allowlists are insufficient, and compliance evidence artifacts for SOC 2, HIPAA, PCI-DSS, FedRAMP, DORA, NIST AI RMF, and EU AI Act. For security architects.

Download PDF →
Solution Brief

Field & Buyer Overview

The threat narrative, why in-harness allowlists sit inside the trust boundary Aviatrix operates outside of, the before/after for AI workflow engineers, compliance proof points, and discovery questions for security and platform buyers.

Download PDF →
GitHub

Full Terraform Blueprint

VPC, private subnet, OpenClaw/NemoClaw VM, SSM-only management, VPC Flow Logs, Aviatrix Spoke Gateway, DCF policy objects, agent-class preset tfvars profiles (locked-down, coding, research, support, regulated), validation tests, and operations docs including AGENTS.md, domain tiers, preflight checks, and rollback guidance.

View repository →
Video

Attack simulation

60-second lab recording. A prompt-injected OpenClaw agent attempts to exfiltrate credentials to an unapproved model provider. The priority-10 shadow-model deny rule fires before the TCP handshake completes. CoPilot FlowIQ logs the attempt with full agent-class identity and rule attribution.

Watch →

Trusted by enterprise security teams

SOC 2 Type II

Independently audited

ISO 27001

Certified

500+ enterprises

Including 10% of the Fortune 500

Zero data-plane access

Aviatrix never touches your traffic

Known constraints

Documented before you find them in production.

Lab-validated limitations and workarounds for the OpenClaw agent runtime containment VCA. Published upfront so your deployment matches the docs.

Reference blueprint — statically reviewed, not live-tested

This architecture has been statically reviewed but not applied against a live AWS account or Aviatrix Controller. Run terraform fmt, validate, plan, and a non-production monitor-mode deployment before production use.

Developer workstations are out of scope

Unmanaged workstation and personal VPS agent runtimes are out of scope. Routing is not deterministic and enforcement cannot sit outside the host. The recommended enterprise posture is to avoid approving unmanaged workstation agents for regulated data or production access.

K8s shape requires Controller 9.0+ for TLS decryption

The VM shape works on Controller 8.1+. The Kubernetes extension requires Controller 9.0+ for selective TLS decryption, URL-path filtering, and egress IDS/IPS. Controller 8.2+ is required for the POST_RULES default-action resource.

Per-agent lifecycle automation is Phase 3

An Agent Egress Identity Service integrating with the enterprise agent registry, Kubernetes admission workflow, or AI platform control plane is a design candidate. This VCA delivers the network enforcement layer; agent lifecycle automation is a future phase.

Rollout principle

Deploy in monitor mode first. Baseline. Then enforce.

Deploy the private VM or Kubernetes namespace through the blueprint with policy_mode = monitor. Exercise real terminal workflows — install, clone, pull packages, call the model gateway, use approved SaaS/MCP tools. Convert observed legitimate destinations into WebGroups by pull request. Switch to policy_mode = enforce, run the egress verification, and make the preset reusable for the next team.

Deploy today

The Terraform is built.
The policy pack is on GitHub.

Deploy the agent runtime in monitor mode this afternoon. Baseline your agent’s real destinations, then enforce. Pick the path that matches where you are.

New to Aviatrix

Start with Enterprise — VCAs included free

Subscribe on AWS Marketplace, deploy Enterprise in under 15 minutes, then deploy this VCA on top. 30-day free trial — VCAs included.

Subscribe on AWS Marketplace · 30-day free trial · No contract

Already on Enterprise

Pull the Terraform from GitHub

Full blueprint, agent-class preset profiles, validation tests, and AGENTS.md. Deploy in monitor mode, baseline, then enforce.

Get the Terraform →

Controller 8.1+ · 1 managed network per deployment · Monitor-first rollout