<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[Nertwork]]></title><description><![CDATA[Nertwork]]></description><link>https://nertwork.hashnode.dev</link><generator>RSS for Node</generator><lastBuildDate>Wed, 18 Nov 2026 15:16:02 GMT</lastBuildDate><atom:link href="https://nertwork.hashnode.dev/rss.xml" rel="self" type="application/rss+xml"/><language><![CDATA[en]]></language><ttl>60</ttl><item><title><![CDATA[Enterprise Networking Trends: SASE, Multi-Cloud, Private 5G, and AI Network Management]]></title><description><![CDATA[Architectural analysis of SASE, zero-trust WAN convergence, multi-cloud interconnects, private 5G, and AI network management.]]></description><link>https://nertwork.hashnode.dev/enterprise-networking-trends-sase-multi-cloud-private-5g-and-ai-network-management</link><guid isPermaLink="true">https://nertwork.hashnode.dev/enterprise-networking-trends-sase-multi-cloud-private-5g-and-ai-network-management</guid><category><![CDATA[networking]]></category><category><![CDATA[Cloud Computing]]></category><category><![CDATA[Security]]></category><category><![CDATA[AI]]></category><category><![CDATA[architecture]]></category><dc:creator><![CDATA[James Farr Gomez]]></dc:creator><pubDate>Tue, 17 Nov 2026 06:20:29 GMT</pubDate><content:encoded><![CDATA[<h2>1. Network and Security Convergence</h2>
<p>In 2023, one of the key networking trends is the convergence of network and security operations. With the rise of distributed IT environments and hybrid work models, networks have become crucial for identifying and mitigating cyberattacks. Organizations are increasingly integrating Zero Trust and Secure Access Service Edge (SASE) frameworks into their network infrastructure. The integration of network and security tools is vital in this process, although it might be a gradual transition due to existing IT investments. Network traffic information is becoming increasingly important for detecting and addressing anomalies, thus bolstering security postures (source: TechTarget).</p>
<h2>2. Multi-cloud Networking</h2>
<p>The use of multiple public clouds has significantly increased, with 86% of organizations reported to be using multiple public clouds for Infrastructure as a Service (IaaS) and Platform as a Service (PaaS). This trend leads to more complex network management but also drives the development of tools that simplify connections to multiple cloud vendors, thereby enhancing organizational agility (source: TechTarget).</p>
<h2>3. Private 5G and Wi-Fi Delineation</h2>
<p>Another notable trend is the growing delineation between private 5G and Wi-Fi use cases. While Wi-Fi continues to dominate enterprise spaces, private 5G deployments are on the rise, especially in industrial settings. The choice between these technologies often depends on the specific application requirements. Private 5G is emerging as a primary connection for some SD-WAN deployments, although it can be complex and costly (source: TechTarget).</p>
<h2>4. Digital Twin Technology</h2>
<p>Digital twins, which create virtual replicas of physical network environments, are seeing increased adoption. These technologies allow for safer testing and optimization of network configurations and operations without affecting the physical network. Such advancements are crucial for enhancing business resiliency and operational efficiency (source: TechTarget and EnterpriseNetworkingPlanet).</p>
<h2>5. AI-based Network Management</h2>
<p>Artificial intelligence (AI) is revolutionizing network management by automating manual processes and improving problem detection and triage. AI-driven tools are increasingly used for proactive network management, enhancing user experiences and reducing downtime (source: EnterpriseNetworkingPlanet).</p>
<h2>6. Next-Generation Connectivity</h2>
<p>Next-generation connectivity encompasses the development and implementation of advanced network technologies, including 5G and next-generation Wi-Fi. These technologies are crucial for enabling high-speed, low-latency connectivity in various settings, from smart homes to large public venues and enterprises (source: EnterpriseNetworkingPlanet).</p>
<h2>Conclusion</h2>
<p>Staying abreast of these trends is crucial for network professionals and organizations looking to optimize their network infrastructure and security. The trends reflect an ongoing evolution towards more integrated, intelligent, and versatile networking solutions that cater to the demands of modern digital enterprises and public services.</p>
<p>For more detailed insights and specific examples, you can refer to the original articles on <a href="https://www.techtarget.com/">TechTarget</a>, <a href="https://www.enterprisenetworkingplanet.com/">EnterpriseNetworkingPlanet</a>, and other sources cited.</p>
<hr />
<p><em>Originally published by</em> <a href="https://www.nertwork.com/about/"><em>James Farr Gomez</em></a> <em>on</em> <a href="https://www.nertwork.com/2023/12/07/current-networking-trends-in-2023-an-in-depth-analysis/"><em>Nertwork</em></a><em>.</em></p>
]]></content:encoded></item><item><title><![CDATA[Enforcing Network Standards Across a 1,000+ Device Fleet with Agentic AI]]></title><description><![CDATA[Enforcing multi-vendor Juniper, Cisco, and Cumulus network standards across 1,000+ devices with NetBox SSOT and agentic AI.]]></description><link>https://nertwork.hashnode.dev/enforcing-network-standards-across-a-1-000-device-fleet-with-agentic-ai</link><guid isPermaLink="true">https://nertwork.hashnode.dev/enforcing-network-standards-across-a-1-000-device-fleet-with-agentic-ai</guid><category><![CDATA[networking]]></category><category><![CDATA[AI]]></category><category><![CDATA[automation]]></category><category><![CDATA[Security]]></category><category><![CDATA[Devops]]></category><dc:creator><![CDATA[James Farr Gomez]]></dc:creator><pubDate>Tue, 17 Nov 2026 06:19:12 GMT</pubDate><content:encoded><![CDATA[<p>Enterprise networks rarely fail from a single bug—they fail from <strong>configuration entropy</strong>:</p>
<ul>
<li>Interface descriptions referencing retired peers (<code>TEMP_UPLINK_DO_NOT_TOUCH</code>)</li>
<li>Drifted hostnames across datacenters, cloud interconnects, and branches</li>
<li>Legacy ACLs &amp; firewall policies with shadowed rules, <code>any</code> services, and stale objects</li>
<li>Undocumented BGP neighbors, VTI tunnels, and LAG members</li>
</ul>
<p>Manual audits take months. Here is how we built an <strong>Agentic AI (</strong><code>ai-ops</code><strong>)</strong> pipeline to enforce deterministic standards across 1,000+ devices—without letting an LLM wing a live commit:</p>
<img src="https://i0.wp.com/www.nertwork.com/wp-content/uploads/2026/10/agentic-network-architecture.png?resize=1600%2C900&amp;ssl=1" alt="Two-Tier Agentic AI Network Standards Pipeline with NetBox SSOT and Oxidized AST Parsing" style="display:block;margin:0 auto" />

<p>Four-stage network standards enforcement pipeline: NetBox/Oxidized correlation, deterministic AST checks, two-tier Agentic AI, and human approval gate.</p>
<h2>1. Multi-Source Truth Correlation (~50ms)</h2>
<p>Before proposing changes, agents correlate 4 live sources:</p>
<ul>
<li><strong>Oxidized:</strong> Config history across firewalls, switches, and routers</li>
<li><strong>NetBox (SSOT):</strong> Authoritative racks, circuits, prefixes, and roles</li>
<li><strong>LibreNMS &amp; Prometheus:</strong> Live LLDP/CDP, ARP/MAC, and BGP state</li>
<li><strong>Graphify Graph:</strong> Pre-compiled topology queried in <code>&lt;150</code> tokens for blast radius</li>
</ul>
<h2>2. Semantic Interface, Hostname &amp; Tunnel Normalization</h2>
<p>Cross-verifying LLDP/CDP and BGP state against IPAM enforces structured schemas:</p>
<ul>
<li><strong>Interfaces:</strong> <code>&lt;peer_host&gt;:&lt;peer_port&gt; | &lt;role/circuit&gt; | &lt;vrf/vlan&gt;</code></li>
<li><strong>Hostnames:</strong> <code>&lt;site&gt;-&lt;role&gt;-&lt;id&gt;</code> aligned across CLI, DNS, IPAM, and monitoring</li>
<li><strong>VTI Tunnels:</strong> Standardized metadata with remote ASN, loopbacks, and routing boundaries (cloud workload vs. management paths)</li>
</ul>
<h2>3. Autonomous ACL &amp; Firewall Policy Hygiene</h2>
<p>Agents parse vendor syntax into normalized models to:</p>
<ul>
<li>Replace inline IPs with canonical, role-scoped address objects</li>
<li>Flag shadowed rules, expired exceptions, and zero-hit policies via hit counters</li>
<li>Enforce least-privilege segmentation across corporate, cloud, and branch zones</li>
</ul>
<h2>4. Orchestrator + Disposable Sub-Agent Architecture</h2>
<p>To prevent LLM context rot at scale, we use a two-tier design:</p>
<ul>
<li><strong>Orchestrator (</strong><code>&lt;40k</code> <strong>tokens):</strong> Plans site waves and synthesizes fleet diffs</li>
<li><strong>Disposable Sub-Agents:</strong> Audit one site/rack, query the subgraph, and output compact diffs—cutting tokens ~80% with 100% precision</li>
</ul>
<h2>5. Candidate Staging &amp; Human-in-the-Loop Approval</h2>
<p>AI never autonomously commits to live gear:</p>
<ol>
<li>Stage in candidate config or dry-run mode</li>
<li>Run syntax, policy, and routing pre-checks</li>
<li>Attach an exact Before/After diff to the change ticket</li>
<li>Require <strong>Human Approval</strong> before commit + post-change BGP/port verification</li>
</ol>
<h2>The Result</h2>
<p>Multi-quarter cleanups are now a continuous pipeline—turning <code>show interface status</code> into a living diagram.</p>
<p><em>How are you tackling config drift and ACL hygiene at scale?</em></p>
<hr />
<p><em>Originally published by</em> <a href="https://www.nertwork.com/about/"><em>James Farr Gomez</em></a> <em>on</em> <a href="https://www.nertwork.com/2026/10/30/enforcing-network-standards-across-a-1000-device-fleet-with-agentic-ai/"><em>Nertwork</em></a><em>.</em></p>
]]></content:encoded></item><item><title><![CDATA[Leveraging AI for a Homelab: Declarative Proxmox VE Infrastructure with Terraform, Pulumi, and GitLab CI/CD]]></title><description><![CDATA[Zero-click Proxmox VE provisioning with Terraform, Pulumi, Ansible, GitLab CI/CD, and deterministic AI ops tooling.]]></description><link>https://nertwork.hashnode.dev/leveraging-ai-for-a-homelab-declarative-proxmox-ve-infrastructure-with-terraform-pulumi-and-gitlab-ci-cd</link><guid isPermaLink="true">https://nertwork.hashnode.dev/leveraging-ai-for-a-homelab-declarative-proxmox-ve-infrastructure-with-terraform-pulumi-and-gitlab-ci-cd</guid><category><![CDATA[Devops]]></category><category><![CDATA[Terraform]]></category><category><![CDATA[Kubernetes]]></category><category><![CDATA[GitLab]]></category><category><![CDATA[AI]]></category><dc:creator><![CDATA[James Farr Gomez]]></dc:creator><pubDate>Tue, 17 Nov 2026 06:19:01 GMT</pubDate><content:encoded><![CDATA[<p>Modern homelabs have evolved far beyond a single hypervisor running a handful of manually clicked-together virtual machines. When your homelab mirrors production enterprise architecture—spanning a multi-node <strong>Proxmox VE</strong> cluster, Kubernetes workloads, VLAN-segmented MikroTik and UniFi switching, HashiCorp Vault secret management, and full observability—manual GUI provisioning (“ClickOps”) quickly leads to configuration drift.</p>
<p>Here is how we architected a 100% declarative, self-healing homelab by combining <strong>Terraform</strong>, <strong>Pulumi</strong>, <strong>GitLab CI/CD</strong>, and <strong>leveraging AI for a homelab</strong> (<code>ai-ops</code>) to manage the entire lifecycle from bare-metal hypervisors to application services.</p>
<img src="https://i0.wp.com/www.nertwork.com/wp-content/uploads/2026/11/proxmox-iac-architecture.png?resize=1600%2C900&amp;ssl=1" alt="Declarative Proxmox VE Homelab IaC Architecture: Terraform, Pulumi &amp; GitLab CI/CD" style="display:block;margin:0 auto" />

<p>Declarative GitOps workflow: GitLab CI/CD plans &amp; drift gates driving Terraform (Proxmox VMs) and Pulumi Python (Kubernetes &amp; Ingress).</p>
<h2>1. Why Declarative IaC + Leveraging AI for a Homelab Matters</h2>
<p>Traditional homelab setups fail in predictable ways: undocumented VM hardware tweaks, snowflake storage mounts, forgotten firewall rules, and stale DNS records. By treating the homelab as an immutable Infrastructure-as-Code (IaC) fleet and <strong>leveraging AI for a homelab</strong> operations plane, we achieve four core guarantees:</p>
<ul>
<li><strong>Zero ClickOps Drift:</strong> Every Proxmox VM, LXC container, VLAN bridge, and DNS record is defined in Git and reconciled via Terraform or Pulumi.</li>
<li><strong>Single Source of Truth (SSOT):</strong> NetBox tracks every rack unit, hypervisor node, VLAN prefix (<code>172.28.x.0/24</code>), and IP assignment before a VM is ever instantiated.</li>
<li><strong>Vault-Backed Ephemeral Secrets:</strong> API tokens, SSH keys, and Cloud-Init credentials are fetched dynamically from HashiCorp Vault at pipeline execution time—never hardcoded in repositories.</li>
<li><strong>Agentic AI Topology Awareness:</strong> Our <code>ai-ops</code> agents query a pre-compiled knowledge graph (<code>graphify</code>) in <code>&lt;150</code> tokens to understand blast radius, author IaC Merge Requests, and validate post-deploy telemetry automatically.</li>
</ul>
<h2>2. Proxmox VE Cluster Foundation &amp; Cloud-Init Golden Templates</h2>
<p>Before Terraform or Pulumi touches the cluster, our Proxmox VE nodes (<code>pve01</code>–<code>pve03</code>) are standardized with ZFS/NVMe storage pools, VLAN-aware Linux bridges (<code>vmbr0</code> trunking production, IoT, lab, and management VLANs), and minimal Cloud-Init golden templates.</p>
<p>Building a reproducible Ubuntu / Debian Cloud-Init template (<code>VMID 9000</code>) on Proxmox takes five CLI commands:</p>
<pre><code class="language-cli">qm create 9000 --name ubuntu-2404-cloudinit --memory 2048 --cores 2 --net0 virtio,bridge=vmbr0
qm importdisk 9000 noble-server-cloudimg-amd64.img local-zfs
qm set 9000 --scsihw virtio-scsi-single --scsi0 local-zfs:vm-9000-disk-0,discard=on,ssd=1
qm set 9000 --ide2 local-zfs:cloudinit --boot c --bootdisk scsi0 --serial0 socket --vga serial0
qm set 9000 --agent enabled=1 &amp;&amp; qm template 9000
</code></pre>
<h2>3. Declarative VM Provisioning with Terraform</h2>
<p>For core stateful infrastructure—such as NetBox, GitLab runners, Wazuh SIEM, and DNS resolvers—we use <strong>Terraform</strong> with the <code>bpg/proxmox</code> provider and HashiCorp Vault integration. Cloning our Cloud-Init template into a production VM with static VLAN tagging and SSH key injection is completely declarative:</p>
<pre><code class="language-cli">terraform {
  required_providers {
    proxmox = {
      source  = "bpg/proxmox"
      version = "~&gt; 0.66.0"
    }
    vault = {
      source  = "hashicorp/vault"
      version = "~&gt; 4.4.0"
    }
  }
}

data "vault_kv_secret_v2" "proxmox_creds" {
  mount = "secret"
  name  = "ai-ops/proxmox"
}

provider "proxmox" {
  endpoint  = data.vault_kv_secret_v2.proxmox_creds.data["api_url"]
  api_token = data.vault_kv_secret_v2.proxmox_creds.data["api_token"]
  insecure  = false
}

resource "proxmox_virtual_environment_vm" "netctl_node" {
  name        = "netctl01"
  node_name   = "pve01"
  vm_id       = 25510
  description = "Managed by Terraform &amp; GitLab CI/CD (ai-ops)"
  tags        = ["terraform", "homelab", "netdevops"]

  cpu {
    cores = 4
    type  = "host"
  }

  memory {
    dedicated = 8192
  }

  clone {
    vm_id = 9000
    full  = true
  }

  disk {
    datastore_id = "local-zfs"
    interface    = "scsi0"
    size         = 64
    discard      = "on"
    ssd          = true
  }

  network_device {
    bridge  = "vmbr0"
    model   = "virtio"
    vlan_id = 255
  }

  initialization {
    ip_config {
      ipv4 {
        address = "172.28.255.10/24"
        gateway = "172.28.255.1"
      }
    }
    user_account {
      username = "ops"
      keys     = [data.vault_kv_secret_v2.proxmox_creds.data["ssh_pubkey"]]
    }
  }
}
</code></pre>
<h2>4. Dynamic Infrastructure &amp; Kubernetes Fleets with Pulumi</h2>
<p>While Terraform excels at static resource blocks, <strong>Pulumi</strong> allows us to use real Python and TypeScript to dynamically query our NetBox SSOT and generate entire fleets of Kubernetes (<code>home1</code>) worker nodes, GPU passthrough VMs, and monitoring probes without duplicating boilerplate HCL:</p>
<pre><code class="language-cli">import pulumi
import pulumi_proxmoxve as proxmox

# Dynamically provision Kubernetes worker nodes across Proxmox cluster hosts
k8s_workers = [
    {"name": "k8s-worker-01", "node": "pve01", "vmid": 25221, "ip": "172.28.252.21/24", "cores": 8, "mem": 16384},
    {"name": "k8s-worker-02", "node": "pve02", "vmid": 25222, "ip": "172.28.252.22/24", "cores": 8, "mem": 16384},
    {"name": "k8s-worker-03", "node": "pve03", "vmid": 25223, "ip": "172.28.252.23/24", "cores": 8, "mem": 16384},
]

for w in k8s_workers:
    proxmox.vm.VirtualMachine(
        w["name"],
        node_name=w["node"],
        vm_id=w["vmid"],
        name=w["name"],
        tags=["pulumi", "kubernetes", "home1"],
        cpu=proxmox.vm.VirtualMachineCpuArgs(cores=w["cores"], type="host"),
        memory=proxmox.vm.VirtualMachineMemoryArgs(dedicated=w["mem"]),
        clone=proxmox.vm.VirtualMachineCloneArgs(vm_id=9000, full=True),
        network_devices=[
            proxmox.vm.VirtualMachineNetworkDeviceArgs(bridge="vmbr0", model="virtio", vlan_id=252)
        ],
        initialization=proxmox.vm.VirtualMachineInitializationArgs(
            ip_configs=[
                proxmox.vm.VirtualMachineInitializationIpConfigArgs(
                    ipv4=proxmox.vm.VirtualMachineInitializationIpConfigIpv4Args(
                        address=w["ip"],
                        gateway="172.28.252.1",
                    )
                )
            ]
        ),
    )
</code></pre>
<h2>5. Automated GitLab CI/CD Pipelines with Human-in-the-Loop Gating</h2>
<p>Every infrastructure change flows through our self-hosted <strong>GitLab CI/CD</strong> instance. Merge Requests automatically trigger syntax validation, security scanning, <code>terraform plan</code>, and <code>pulumi preview</code>, posting the exact infrastructure diff back to the Merge Request before a human approves the production apply:</p>
<pre><code class="language-cli">stages:
  - validate
  - plan
  - apply

variables:
  TF_IN_AUTOMATION: "true"

terraform:validate:
  stage: validate
  image: hashicorp/terraform:1.9
  script:
    - terraform init -backend=false
    - terraform fmt -check
    - terraform validate

terraform:plan:
  stage: plan
  image: hashicorp/terraform:1.9
  script:
    - terraform init
    - terraform plan -out=tfplan.binary
    - terraform show -no-color tfplan.binary &gt; plan.txt
  artifacts:
    paths:
      - tfplan.binary
      - plan.txt
    expire_in: 7 days

terraform:apply:
  stage: apply
  image: hashicorp/terraform:1.9
  script:
    - terraform init
    - terraform apply -auto-approve tfplan.binary
  dependencies:
    - terraform:plan
  rules:
    - if: $CI_COMMIT_BRANCH == "master"
      when: manual
</code></pre>
<h2>6. Leveraging AI for a Homelab: The Agentic AI Control Plane</h2>
<p>The true multiplier in our architecture is <strong>leveraging AI for a homelab</strong> across day-2 operations. Instead of giving an LLM unconstrained root access, we pair an <strong>Orchestrator + Scoped Sub-Agent architecture</strong> with deterministic guardrails:</p>
<ol>
<li><strong>Knowledge Graph &amp; Topology Context (</strong><code>&lt;150</code> <strong>Tokens):</strong> Before proposing any Proxmox, Kubernetes, or network change, agents query our NetBox-backed topology CLI and Graphify knowledge graph to map dependencies across hypervisors, VLANs, and services.</li>
<li><strong>Automated Drift &amp; CI Triage:</strong> When a GitLab CI/CD pipeline fails or Prometheus/Wazuh fires an alert, an autonomous remediation agent inspects the logs, correlates recent commits, and stages a targeted Merge Request.</li>
<li><strong>Maintenance Window Coordination:</strong> Before restarting Kubernetes workloads or migrating Proxmox VMs, agents pause auto-remediation watchdogs so health probes never fight an active deployment.</li>
<li><strong>Post-Change Telemetry Verification:</strong> After a Terraform or Pulumi apply completes, agents verify QEMU guest agent heartbeats, Prometheus node-exporter metrics, and Blackbox HTTPS probes to confirm 100% service health.</li>
</ol>
<p>By combining declarative Proxmox IaC (<strong>Terraform + Pulumi + GitLab CI/CD</strong>) with <strong>Agentic AI operations</strong>, our homelab runs with the reliability, auditability, and velocity of a hyperscale cloud environment.</p>
<hr />
<p><em>Originally published by</em> <a href="https://www.nertwork.com/about/"><em>James Farr Gomez</em></a> <em>on</em> <a href="https://www.nertwork.com/2026/11/10/leveraging-ai-for-a-homelab-proxmox-terraform-gitlab-pulumi/"><em>Nertwork</em></a><em>.</em></p>
]]></content:encoded></item><item><title><![CDATA[Autonomous Cloud & Infrastructure SecOps: Integrating Wazuh SIEM/XDR, HashiCorp Vault, and Agentic AI Across AWS, GCP, and Proxmox Fleets]]></title><description><![CDATA[Unifying Wazuh SIEM/XDR, HashiCorp Vault, and deterministic agentic AI triage across AWS, GCP, and Proxmox infrastructure fleets.]]></description><link>https://nertwork.hashnode.dev/autonomous-cloud-infrastructure-secops-integrating-wazuh-siem-xdr-hashicorp-vault-and-agentic-ai-across-aws-gcp-and-proxmox-fleets</link><guid isPermaLink="true">https://nertwork.hashnode.dev/autonomous-cloud-infrastructure-secops-integrating-wazuh-siem-xdr-hashicorp-vault-and-agentic-ai-across-aws-gcp-and-proxmox-fleets</guid><category><![CDATA[Security]]></category><category><![CDATA[Devops]]></category><category><![CDATA[AWS]]></category><category><![CDATA[GCP]]></category><category><![CDATA[AI]]></category><dc:creator><![CDATA[James Farr Gomez]]></dc:creator><pubDate>Tue, 17 Nov 2026 06:18:51 GMT</pubDate><content:encoded><![CDATA[<p>Modern enterprise infrastructure rarely lives inside a single trust boundary. A typical production environment spans <strong>Google Cloud Platform (GCP)</strong> GKE clusters and Shared VPCs, <strong>Amazon Web Services (AWS)</strong> Transit Gateway spokes and EC2 workloads, perimeter <strong>Juniper SRX / Cisco ASA</strong> security zones, and on-premises <strong>Proxmox VE</strong> and <strong>Kubernetes</strong> bare-metal nodes. While open-source <strong>Wazuh SIEM &amp; XDR</strong> provides exceptional unified visibility across all of these environments, traditional security operations workflows break down at scale for two reasons: <strong>alert fatigue</strong> and <strong>LLM context-window bloat</strong>.</p>
<p>If you feed raw OpenSearch JSON alert documents or unfiltered vulnerability indices directly into an AI coding or operations agent, a single query can consume 40,000+ tokens—causing high latency, hallucinated package names, and missed stale-index edge cases. In this architectural deep dive, we walk through how we engineered a <strong>zero-dependency, sub-150-token Wazuh SIEM/XDR CLI and Model Context Protocol (MCP) bridge</strong> backed by <strong>HashiCorp Vault</strong>, and how our two-tier <strong>Agentic AI (</strong><code>ai-ops</code><strong>)</strong> platform correlates hybrid cloud telemetry (AWS GuardDuty/CloudTrail, GCP Cloud Audit Logs, and Linux/Kubernetes XDR agents) into deterministic <strong>Ansible and Terraform</strong> remediation pipelines.</p>
<img src="https://i0.wp.com/www.nertwork.com/wp-content/uploads/2026/11/wazuh-secops-architecture.png?ssl=1" alt="Autonomous Multi-Cloud SecOps Architecture: Wazuh SIEM/XDR, HashiCorp Vault &amp; Agentic AI" style="display:block;margin:0 auto" />

<p>End-to-end multi-cloud SecOps pipeline: AWS/GCP/Proxmox telemetry normalized via Wazuh &amp; HashiCorp Vault into deterministic Agentic AI remediation.</p>
<h2>1. Why Traditional SIEM Workflows Fail Agentic AI</h2>
<p>Wazuh v4.x separates security operations across two distinct API planes:</p>
<ul>
<li><strong>The Wazuh Server REST API (</strong><code>55000/tcp</code><strong>):</strong> Manages JWT authentication (<code>/security/user/authenticate</code>), agent enrollment and health (<code>/agents</code>), CIS Security Configuration Assessment (SCA) pass/fail ratios (<code>/sca/&lt;agent_id&gt;</code>), and active-response firewall blocks.</li>
<li><strong>The OpenSearch Indexer API (</strong><code>9200/tcp</code><strong>):</strong> Stores real-time MITRE ATT&amp;CK security alerts in <code>wazuh-alerts-4.x-*</code> and per-package CVE vulnerability states in <code>wazuh-states-vulnerabilities-*</code>.</li>
</ul>
<p>When querying <code>wazuh-states-vulnerabilities-*</code> across a hybrid fleet, we discovered three operational traps that naive dashboards and scripts miss:</p>
<ol>
<li><strong>Disconnected Ephemeral Cloud Nodes:</strong> Auto-scaled AWS EC2 or GCP GKE nodes that terminated days ago leave orphaned CVE documents in the OpenSearch vulnerability index unless filtered against live <code>status=active</code> agent IDs from port <code>55000</code>.</li>
<li><strong>Stale Pre-Patch Index Records:</strong> After an Ansible kernel or package upgrade, historical vulnerability documents with older <code>vulnerability.detected_at</code> timestamps can persist until the next full syscollector scan or explicit <code>_delete_by_query</code> index hygiene pass.</li>
<li><strong>Multi-Megabyte JSON Responses:</strong> Querying 500 alerts or CVEs without server-side OpenSearch aggregations (<code>terms</code>, <code>cardinality</code>, <code>top_hits</code>) dumps megabytes of verbose JSON that overwhelm both human operators and AI orchestrators.</li>
</ol>
<p>| SecOps Capability | Raw Wazuh / OpenSearch API | Deterministic AI-Ops CLI (<code>wazuh_cli.py</code>) |<br />| <strong>Executive Fleet Posture</strong> | 42,000+ tokens across 6 REST &amp; OpenSearch calls | <strong>&lt;140 tokens</strong> (Active agents, deduplicated CVEs, CIS SCA score, top MITRE rules) |<br />| <strong>Secret &amp; Token Resolution</strong> | Static plaintext <code>.env</code> files or manual dashboard login | <strong>HashiCorp Vault KV v2</strong> dynamic resolution + ephemeral JWT caching |<br />| <strong>Stale &amp; Disconnected Filtering</strong> | Manual cross-referencing between Indexer and Manager | Automatic active-agent intersection + <code>_delete_by_query</code> stale index pruning |<br />| <strong>Remediation Execution</strong> | Manual SSH package updates or ad-hoc shell scripts | Categorized <strong>Ansible / Terraform GitOps</strong> playbooks with maintenance-window locks |</p>
<h2>2. Unifying AWS, GCP, Network Perimeter &amp; Proxmox Telemetry</h2>
<p>To eliminate blind spots between cloud control planes and on-premises hypervisors, every security domain streams structured events into our central Wazuh XDR pipeline:</p>
<ul>
<li><strong>Amazon Web Services (AWS):</strong> AWS <strong>CloudTrail</strong> management/data events, <strong>GuardDuty</strong> threat findings, and <strong>VPC Flow Logs</strong> stream via S3/Kinesis into the Wazuh AWS module (<code>wodle name="aws-s3"</code>), correlating IAM privilege escalations (<code>rule.level &gt;= 10</code>) alongside host-level EC2 and EKS container telemetry.</li>
<li><strong>Google Cloud Platform (GCP):</strong> GCP <strong>Cloud Audit Logs</strong>, <strong>Cloud Armor</strong> WAF blocks, and <strong>GKE</strong> audit streams route through a Google Cloud Pub/Sub subscription (<code>wodle name="gcp-pubsub"</code>) authenticated via short-lived Workload Identity credentials.</li>
<li><strong>Perimeter Network Security (</strong><code>JNCIP-SEC</code><strong>):</strong> <strong>Juniper SRX</strong> security zones, <strong>Cisco ASA/Firepower</strong> appliances, and edge routers stream structured syslog (<code>514/udp</code> / TLS) into custom Wazuh decoders that trigger automated edge active-response drops when brute-force or port-scan thresholds are exceeded.</li>
<li><strong>Proxmox VE &amp; Kubernetes Nodes:</strong> Lightweight Wazuh agents on every Debian/Ubuntu hypervisor and Kubernetes worker run continuous file integrity monitoring (<code>syscheck</code>), package/kernel inventory (<code>syscollector</code>), and <strong>CIS Benchmark Security Configuration Assessments (</strong><code>sca</code><strong>)</strong>.</li>
</ul>
<pre><code class="language-yaml"># /var/ossec/etc/ossec.conf — Hybrid AWS, GCP &amp; CIS Benchmark Telemetry Configuration
ossec_config:
  wodles:
    - aws-s3:
        disabled: "no"
        interval: "5m"
        buckets:
          - name: "enterprise-secops-cloudtrail-logs"
            type: "cloudtrail"
            regions: "us-west-2,us-east-1,eu-central-1"
          - name: "enterprise-secops-guardduty-findings"
            type: "guardduty"
    - gcp-pubsub:
        project_id: "prod-secops-telemetry-01"
        subscription_name: "wazuh-cloud-audit-and-armor-sub"
        max_messages: 500
        interval: "2m"
  sca:
    enabled: "yes"
    scan_on_start: "yes"
    interval: "12h"
    policies:
      - "etc/shared/cis_ubuntu24-04.yml"
      - "etc/shared/cis_debian12.yml"
</code></pre>
<h2>3. Building the Sub-150-Token Wazuh + HashiCorp Vault CLI</h2>
<p>In our two-tier Agentic AI architecture, the primary orchestrator never holds raw database passwords or parses multi-megabyte JSON payloads. Instead, our zero-dependency Python CLI (<code>wazuh_cli.py</code>) retrieves short-lived credentials from <strong>HashiCorp Vault</strong> (<code>secret/data/ai-ops/wazuh</code>), authenticates to both the Wazuh Manager API (<code>55000/tcp</code>) and the OpenSearch Indexer (<code>9200/tcp</code>), and cross-filters open vulnerabilities against currently active agents:</p>
<pre><code class="language-python">#!/usr/bin/env python3
"""Zero-dependency Wazuh SIEM/XDR + OpenSearch CVE &amp; CIS SCA Query Engine."""
import json
import ssl
import urllib.request

def fetch_active_agent_cves(indexer_url: str, active_agent_ids: list[str], auth_header: str) -&gt; dict:
    """Query OpenSearch wazuh-states-vulnerabilities-* for active agents only."""
    query_dsl = {
        "size": 0,
        "query": {
            "bool": {
                "filter": [
                    {"terms": {"agent.id": active_agent_ids}},
                    {"terms": {"vulnerability.severity": ["Critical", "High"]}},
                ]
            }
        },
        "aggs": {
            "by_severity": {"terms": {"field": "vulnerability.severity", "size": 5}},
            "by_agent": {
                "terms": {"field": "agent.name", "size": 25},
                "aggs": {
                    "top_cves": {
                        "terms": {"field": "vulnerability.id", "size": 10},
                        "aggs": {
                            "pkg": {"terms": {"field": "package.name", "size": 3}}
                        }
                    }
                }
            }
        }
    }
    req = urllib.request.Request(
        f"{indexer_url}/wazuh-states-vulnerabilities-*/_search",
        data=json.dumps(query_dsl).encode("utf-8"),
        headers={"Authorization": auth_header, "Content-Type": "application/json"},
        method="POST",
    )
    ctx = ssl.create_default_context()
    ctx.check_hostname = False
    ctx.verify_mode = ssl.CERT_NONE
    with urllib.request.urlopen(req, context=ctx, timeout=15) as resp:
        return json.loads(resp.read().decode("utf-8"))
</code></pre>
<p>When an engineer or AI sub-agent runs <code>wazuh_cli.py summary</code> or <code>wazuh_cli.py remediate</code>, the CLI classifies every finding into five deterministic buckets:</p>
<ol>
<li><strong>Bucket 1 — Stale OpenSearch Documents:</strong> Automatically pruned via <code>POST /wazuh-states-vulnerabilities-*/_delete_by_query</code> when the package is already upgraded on the host.</li>
<li><strong>Bucket 2 — Disconnected / Terminated Cloud Nodes:</strong> Filtered out of active posture calculations and queued for agent deregistration.</li>
<li><strong>Bucket 3 — Unbooted / Old Linux Kernels:</strong> Hosts where <code>apt-get dist-upgrade</code> installed a patched kernel image, but an older <code>linux-image-*</code> package remains on disk awaiting <code>apt-get autoremove --purge</code>.</li>
<li><strong>Bucket 4 — Container &amp; Ingress Images:</strong> Correlated against upstream registries (Docker Hub / GHCR / GitLab Registry) to verify that a patched container tag actually exists before triggering a GitOps bump.</li>
<li><strong>Bucket 5 — Standard OS Packages (</strong><code>apt</code> <strong>/</strong> <code>dnf</code><strong>):</strong> Grouped into a targeted Ansible remediation batch.</li>
</ol>
<h2>4. Closed-Loop Autonomous Remediation with Ansible, Terraform &amp; GitLab CI/CD</h2>
<p>Detecting a CVE or CIS Benchmark failure is only half the battle. To close the loop safely without risking unplanned outages on stateful hypervisors or Kubernetes control planes, our autonomous remediation workflow enforces three strict guardrails:</p>
<ol>
<li><strong>Maintenance-Window Pause (</strong><code>hermes-ctl pause</code><strong>):</strong> Before touching any monitored service, the remediation pipeline pauses autonomous peer-healers so rolling restarts do not trigger false-positive auto-heal rollbacks.</li>
<li><strong>Ansible Package &amp; Kernel Hygiene Playbook:</strong> Executes non-interactive <code>DEBIAN_FRONTEND=noninteractive</code> security upgrades, removes superseded kernel packages, and restarts affected systemd units only when required.</li>
<li><strong>Post-Patch Index Verification:</strong> Triggers a fresh Wazuh syscollector scan, prunes stale vulnerability documents, and verifies that the fleet returns to <strong>0 Critical / 0 High active CVEs</strong>.</li>
</ol>
<pre><code class="language-yaml"># playbooks/patch_secops_vulns.yml — Deterministic Fleet CVE &amp; Kernel Remediation
- name: Remediate Active Wazuh CVEs and Prune Superseded Kernels
  hosts: secops_patch_targets
  become: true
  serial: 2
  tasks:
    - name: Update apt cache and apply security upgrades non-interactively
      ansible.builtin.apt:
        update_cache: true
        upgrade: dist
        autoremove: true
        purge: true
        dpkg_options: "force-confdef,force-confold"
      environment:
        DEBIAN_FRONTEND: noninteractive

    - name: Restart Wazuh agent to trigger immediate syscollector rescan
      ansible.builtin.systemd:
        name: wazuh-agent
        state: restarted
</code></pre>
<h2>5. Key Takeaways for Cloud &amp; Security Architects</h2>
<ul>
<li><strong>Compress Telemetry Before It Reaches the LLM:</strong> Replacing raw SIEM JSON dumps with a deterministic <code>&lt;150-token</code> CLI reduces AI triage latency from 30+ seconds to under 800 milliseconds while eliminating hallucinations.</li>
<li><strong>Unify Cloud and Bare-Metal Signals:</strong> Correlating <strong>AWS GuardDuty/CloudTrail</strong>, <strong>GCP Cloud Audit Logs/Cloud Armor</strong>, <strong>Juniper SRX / Cisco firewalls</strong>, and <strong>Proxmox/Kubernetes</strong> host telemetry inside Wazuh gives security architects a single queryable graph across every environment.</li>
<li><strong>Never Auto-Patch Without Upstream &amp; Blast-Radius Verification:</strong> Pairing Wazuh vulnerability indices with <strong>NetBox SSOT</strong>, <strong>HashiCorp Vault</strong>, and gated <strong>Ansible/Terraform CI/CD pipelines</strong> turns continuous compliance from a quarterly fire drill into a deterministic daily engineering loop.</li>
</ul>
<hr />
<p><em>Originally published by</em> <a href="https://www.nertwork.com/about/"><em>James Farr Gomez</em></a> <em>on</em> <a href="https://www.nertwork.com/2026/11/10/autonomous-cloud-and-infrastructure-secops-wazuh-siem-xdr-hashicorp-vault-and-agentic-ai/"><em>Nertwork</em></a><em>.</em></p>
]]></content:encoded></item><item><title><![CDATA[Introducing BluePill v1.1.0: Native Qt6 C++/QML iMessage Client for Linux & Zero-Electron macOS Relay]]></title><description><![CDATA[Native C++20 and Qt6/QML iMessage, SMS, and RCS desktop client for Linux and Wayland paired with a zero-Electron macOS relay.]]></description><link>https://nertwork.hashnode.dev/introducing-bluepill-v1-1-0-native-qt6-c-qml-imessage-client-for-linux-zero-electron-macos-relay</link><guid isPermaLink="true">https://nertwork.hashnode.dev/introducing-bluepill-v1-1-0-native-qt6-c-qml-imessage-client-for-linux-zero-electron-macos-relay</guid><category><![CDATA[Linux]]></category><category><![CDATA[cpp]]></category><category><![CDATA[Qt]]></category><category><![CDATA[open source]]></category><category><![CDATA[macOS]]></category><dc:creator><![CDATA[James Farr Gomez]]></dc:creator><pubDate>Tue, 17 Nov 2026 06:16:56 GMT</pubDate><content:encoded><![CDATA[<p>Today, we are thrilled to announce the official release of <strong>BluePill v1.1.0</strong>: a native Qt6 C++ / QML desktop iMessage client for Linux and a zero-Electron, ultra-lightweight macOS relay server. BluePill delivers seamless, end-to-end Apple iMessage connectivity to Linux workstations with near-zero resource overhead, native Wayland / Hyprland integration, and rich messaging features.</p>
<h2>Why We Built BluePill: Escaping Electron Bloat</h2>
<p>For Linux power users and developers, staying connected to Apple ecosystem communication has historically meant running heavy Electron or Flutter bridges like BlueBubbles. While functional, BlueBubbles carries a massive footprint: 400MB to 700MB of RAM, multi-second cold startup times, sluggish UI scrolling, and frequent sync stalls when processing multi-gigabyte message databases.</p>
<p>We designed <strong>BluePill</strong> from first principles to be the high-performance, native alternative:</p>
<ul>
<li><strong>Native Qt6 C++ / QML Desktop Client:</strong> Under 35 MB of RAM, 60fps buttery-smooth hardware-accelerated rendering on modern Linux Wayland (Hyprland, Sway, GNOME, KDE) and X11 desktops. Instant sub-second launch.</li>
<li><strong>Zero-Electron macOS Relay:</strong> A tiny, native Python service running on Big Sur through macOS Tahoe that reads <code>chat.db</code> directly via high-speed SQLite queries and uses AppleScript / native helpers to dispatch messages. Less than 15 MB RAM on macOS.</li>
<li><strong>Direct SSH Transport:</strong> End-to-end encrypted transport over native SSH tunnels (<code>ssh://user@mac-host</code>) with zero port forwarding, no third-party cloud servers, and no reliance on external Firebase/Google push notification servers.</li>
<li><strong>Embedded High-Speed Web UI:</strong> A full-featured web client hosted directly at <code>/web</code> with live 28-bar voice memo waveform rendering, drag-and-drop attachments, and inline tapbacks.</li>
</ul>
<img src="https://i0.wp.com/www.nertwork.com/wp-content/uploads/2026/11/bluepill-v1-1-0-linux-qt6-chat-conversation.png?resize=1440%2C900&amp;ssl=1" alt="BluePill v1.1.0 Linux Qt6/QML iMessage Client Conversation View with Pinned Avatars and Multi-Photo Gallery" style="display:block;margin:0 auto" />

<p>BluePill v1.1.0 native Qt6/QML Linux client displaying Pinned Conversations, inline multi-photo gallery, tapback badges, and rich link previews under Wayland/Hyprland.</p>
<h2>Key Features in v1.1.0</h2>
<ul>
<li><strong>Pinned Conversations &amp; 1-Click 2FA OTP Copy Pill:</strong> Circular avatar row synced with macOS <code>com.apple.messages.pinning.plist</code> plus automatic 4–8 digit verification code extraction.</li>
<li><strong>Inline 28-Bar Voice Memo Waveforms:</strong> Full interactive voice playback with FFmpeg amplitude extraction directly inside the chat view.</li>
<li><strong>Expressive Send &amp; Tapbacks:</strong> Send and receive heart, thumbs up, thumbs down, haha, exclamation, and question tapbacks, plus message effects (Confetti, Slam, Invisible Ink).</li>
<li><strong>Scheduled Messages &amp; Clipboard Screenshot Paste:</strong> Queue messages locally or on the server with arbitrary future timestamps, and press <code>Ctrl+V</code> to paste Wayland/X11 clipboard screenshots directly into any thread.</li>
<li><strong>System Keyring &amp; Waybar Module:</strong> Secure credential storage via Freedesktop <code>SecretService</code> (<code>secret-tool</code> / KWallet) and native Waybar status bar indicators.</li>
<li><strong>Find My Integration:</strong> Query Find My devices and friend locations directly via the REST and RPC APIs.</li>
</ul>
<img src="https://i0.wp.com/www.nertwork.com/wp-content/uploads/2026/11/bluepill-v1-1-0-linux-qt6-home-dashboard.png?resize=1440%2C900&amp;ssl=1" alt="BluePill v1.1.0 Home Dashboard with Pinned Conversations, Sync Window Controls, and Embedded Web UI Launcher" style="display:block;margin:0 auto" />

<p>BluePill v1.1.0 Home Dashboard showing active macOS Relay status, message history sync window controls, and one-click Embedded Web UI launcher.</p>
<h2>Getting Started</h2>
<h3>1. Install macOS Relay Server</h3>
<p>On your Mac (Intel or Apple Silicon):</p>
<pre><code class="language-bash"># Install via Homebrew Tap
brew install nertwork/tap/bluepill

# Start server interactively or as a LaunchAgent
bluepill-server --port 1234 --password my_secret_token
</code></pre>
<h3>2. Install Linux Desktop Client</h3>
<p>On Arch Linux / Omarchy via AUR:</p>
<pre><code class="language-bash">yay -S bluepill
</code></pre>
<p>Or build from source on Debian, Ubuntu, or Fedora using CMake:</p>
<pre><code class="language-bash">cmake -B build -DCMAKE_BUILD_TYPE=Release
cmake --build build -j
sudo cmake --install build
</code></pre>
<h2>Open Source &amp; Autonomous AI Playbooks</h2>
<p>BluePill is 100% open source under the MIT license. We have published full documentation, architecture diagrams, and deterministic AI Agent playbooks for automated setup with Claude Code, Gemini CLI, Cursor, and Codex:</p>
<ul>
<li><strong>GitHub Repository:</strong> <a href="https://github.com/nertwork/bluepill">github.com/nertwork/bluepill</a></li>
<li><strong>Documentation &amp; Web Portal:</strong> <a href="https://nertwork.github.io/bluepill/">nertwork.github.io/bluepill</a></li>
<li><strong>Arch AUR Package:</strong> <a href="https://aur.archlinux.org/packages/bluepill">aur.archlinux.org/packages/bluepill</a></li>
<li><strong>macOS Homebrew Tap:</strong> <a href="https://github.com/nertwork/homebrew-tap">github.com/nertwork/homebrew-tap</a></li>
</ul>
<hr />
<p><em>Originally published by</em> <a href="https://www.nertwork.com/about/"><em>James Farr Gomez</em></a> <em>on</em> <a href="https://www.nertwork.com/2026/11/11/introducing-bluepill-imessage-linux/"><em>Nertwork</em></a><em>.</em></p>
]]></content:encoded></item></channel></rss>