https://samyn92.github.io/agentops/docs/getting-started/Recent content in Getting Started on AgentOpsHugoenhttps://samyn92.github.io/agentops/docs/getting-started/quickstart/Mon, 01 Jan 0001 00:00:00 +0000https://samyn92.github.io/agentops/docs/getting-started/quickstart/<p>This guide gets you from zero to a running AI agent on Kubernetes in three steps.</p> <h2 id="prerequisites">Prerequisites</h2> <ul> <li>Kubernetes cluster <strong>v1.28+</strong> (k3s, kind, EKS, GKE, AKS all work)</li> <li><strong>Helm 3.12+</strong></li> <li><strong>kubectl</strong> configured for your cluster</li> <li>An LLM API key (Anthropic, OpenAI, or Google)</li> </ul> <h2 id="step-1-install-the-platform">Step 1: Install the platform</h2> <div class="highlight"><pre tabindex="0" class="chroma"><code class="language-bash" data-lang="bash"><span class="line"><span class="cl">helm install agentops oci://ghcr.io/samyn92/charts/agentops-platform <span class="se">\ </span></span></span><span class="line"><span class="cl"> --namespace agent-system --create-namespace </span></span></code></pre></div><p>This deploys the operator, console, memory service, and Tempo into <code>agent-system</code>, and creates the <code>agents</code> namespace for workloads.</p>https://samyn92.github.io/agentops/docs/getting-started/installation/Mon, 01 Jan 0001 00:00:00 +0000https://samyn92.github.io/agentops/docs/getting-started/installation/<p>This guide covers installation from prerequisites through custom configuration, verification, and uninstalling.</p> <h2 id="prerequisites">Prerequisites</h2> <table> <thead> <tr> <th>Requirement</th> <th>Minimum version</th> </tr> </thead> <tbody> <tr> <td>Kubernetes</td> <td>1.28+</td> </tr> <tr> <td>Helm</td> <td>3.12+</td> </tr> <tr> <td>kubectl</td> <td>1.28+</td> </tr> <tr> <td>Container runtime</td> <td>containerd, CRI-O</td> </tr> <tr> <td>Storage</td> <td>Default StorageClass with dynamic provisioning (for memory + Tempo PVCs)</td> </tr> </tbody> </table> <p><strong>Required secrets</strong> (created in the <code>agents</code> namespace):</p> <ul> <li>At least one LLM API key (Anthropic, OpenAI, or Google) in a Kubernetes Secret</li> </ul> <h2 id="full-install-all-components">Full install (all components)</h2> <p>The default install deploys the operator, console, memory service, and Tempo:</p>https://samyn92.github.io/agentops/docs/getting-started/architecture/Mon, 01 Jan 0001 00:00:00 +0000https://samyn92.github.io/agentops/docs/getting-started/architecture/<p>AgentOps separates the platform control plane from user agent workloads using a two-namespace model. The operator manages the full lifecycle of agents, tools, resources, and channels as native Kubernetes workloads.</p> <h2 id="component-overview">Component overview</h2> <img src="https://samyn92.github.io/agentops/images/architecture.svg" alt="Platform Architecture" /> <h2 id="two-namespace-model">Two-namespace model</h2> <table> <thead> <tr> <th>Namespace</th> <th>Purpose</th> <th>Contents</th> </tr> </thead> <tbody> <tr> <td><code>agent-system</code></td> <td>Platform control plane</td> <td>Operator, console, Tempo</td> </tr> <tr> <td><code>agents</code></td> <td>User workloads</td> <td>Agent pods, memory service, AgentTool/AgentResource/Channel workloads, PVCs</td> </tr> </tbody> </table> <p>This separation provides:</p> <ul> <li><strong>RBAC isolation</strong> — the operator has broad permissions in <code>agents</code> but users interact only through CRDs</li> <li><strong>Resource quota boundaries</strong> — platform overhead is independent of agent workload limits</li> <li><strong>Clear ownership</strong> — platform upgrades don&rsquo;t disrupt running agents</li> </ul> <h2 id="custom-resource-definitions">Custom Resource Definitions</h2> <p>AgentOps defines 6 CRDs in the <code>agents.agentops.io/v1alpha1</code> API group:</p>