<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>User Guides on Kubemoot</title><link>https://kubemoot.org/docs/user-guides/</link><description>Recent content in User Guides on Kubemoot</description><generator>Hugo</generator><language>en</language><atom:link href="https://kubemoot.org/docs/user-guides/index.xml" rel="self" type="application/rss+xml"/><item><title>The Starter Crew</title><link>https://kubemoot.org/docs/user-guides/starter-crew/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://kubemoot.org/docs/user-guides/starter-crew/</guid><description>&lt;p&gt;&lt;code&gt;kmctl create&lt;/code&gt; scaffolds a small, complete crew you can install on a fresh Kubemoot
cluster and talk to within minutes. It is the Kubemoot equivalent of the nginx chart
&lt;code&gt;helm create&lt;/code&gt; produces: a working example you change into your own crew.&lt;/p&gt;
&lt;p&gt;The starter crew is a read-only guide to the Kubernetes namespace it is installed into.
Ask it what is running, what is wrong, and why. It reads the namespace with real tools and
answers from what it found. It never changes anything.&lt;/p&gt;</description></item><item><title>Build a Crew</title><link>https://kubemoot.org/docs/user-guides/build-a-crew/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://kubemoot.org/docs/user-guides/build-a-crew/</guid><description>&lt;p&gt;This guide builds a crew from its parts. A crew is a set of Kubernetes resources, so
&amp;ldquo;building&amp;rdquo; one means declaring those resources - usually packaged together as a Helm
chart so they version and deploy as a unit.&lt;/p&gt;
&lt;p&gt;To see a complete small crew before writing your own, scaffold the
&lt;a href="../starter-crew/"&gt;starter crew&lt;/a&gt; with &lt;code&gt;kmctl create&lt;/code&gt;.&lt;/p&gt;
&lt;h2 id="what-a-crew-is-made-of"&gt;What a crew is made of&lt;/h2&gt;
&lt;table&gt;
 &lt;thead&gt;
 &lt;tr&gt;
 &lt;th&gt;Resource&lt;/th&gt;
 &lt;th&gt;Role in the crew&lt;/th&gt;
 &lt;/tr&gt;
 &lt;/thead&gt;
 &lt;tbody&gt;
 &lt;tr&gt;
 &lt;td&gt;&lt;code&gt;Crew&lt;/code&gt;&lt;/td&gt;
 &lt;td&gt;The team - groups agents by the &lt;code&gt;kubemoot.ai/crew&lt;/code&gt; label, enables the discussion gateway, configures working memory.&lt;/td&gt;
 &lt;/tr&gt;
 &lt;tr&gt;
 &lt;td&gt;&lt;code&gt;Agent&lt;/code&gt;&lt;/td&gt;
 &lt;td&gt;Each participant: one coordinator, plus Toolers (and optionally Analysts). Declares capabilities, tools, RAG, and prompt refs.&lt;/td&gt;
 &lt;/tr&gt;
 &lt;tr&gt;
 &lt;td&gt;&lt;code&gt;PromptModule&lt;/code&gt;&lt;/td&gt;
 &lt;td&gt;Each agent&amp;rsquo;s behavior, written in &lt;a href="../write-agents-and-adl/"&gt;ADL&lt;/a&gt; and composed by reference.&lt;/td&gt;
 &lt;/tr&gt;
 &lt;tr&gt;
 &lt;td&gt;&lt;code&gt;Model&lt;/code&gt; / &lt;code&gt;ModelProvider&lt;/code&gt;&lt;/td&gt;
 &lt;td&gt;The models (by capability label) and the GPU-backed endpoints that serve them.&lt;/td&gt;
 &lt;/tr&gt;
 &lt;tr&gt;
 &lt;td&gt;&lt;code&gt;MCPServer&lt;/code&gt; / &lt;code&gt;MCPGateway&lt;/code&gt;&lt;/td&gt;
 &lt;td&gt;The tools Toolers use, reached through the gateway.&lt;/td&gt;
 &lt;/tr&gt;
 &lt;tr&gt;
 &lt;td&gt;&lt;code&gt;RAGSource&lt;/code&gt;&lt;/td&gt;
 &lt;td&gt;Knowledge sources, indexed and made queryable.&lt;/td&gt;
 &lt;/tr&gt;
 &lt;/tbody&gt;
&lt;/table&gt;
&lt;h2 id="steps"&gt;Steps&lt;/h2&gt;
&lt;h3 id="1-define-the-crew"&gt;1. Define the crew&lt;/h3&gt;
&lt;p&gt;Declare a &lt;code&gt;Crew&lt;/code&gt; with the discussion gateway enabled, and a label every agent will
share. See the &lt;a href="../../reference/crew/"&gt;Crew CRD reference&lt;/a&gt; for the full spec including
the working-memory policy.&lt;/p&gt;</description></item><item><title>Compose a Crew</title><link>https://kubemoot.org/docs/user-guides/compose-a-crew/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://kubemoot.org/docs/user-guides/compose-a-crew/</guid><description>&lt;p&gt;A crew is only as useful as its design. This guide covers the decisions that determine
whether a crew&amp;rsquo;s coordinator routes questions intelligently and whether its Toolers and
Analysts can answer them: which agents to include, how to write resumes the coordinator
can reason over, how to ground the coordinator with broad knowledge and a well-crafted
advisory brief, and how to give each Tooler the narrowly-scoped tools it needs and
each Analyst the focused RAG knowledge it reasons over.&lt;/p&gt;</description></item><item><title>Write Agents &amp; ADL</title><link>https://kubemoot.org/docs/user-guides/write-agents-and-adl/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://kubemoot.org/docs/user-guides/write-agents-and-adl/</guid><description>&lt;p&gt;An agent&amp;rsquo;s behavior is its prompt, and in Kubemoot every prompt is written in
&lt;strong&gt;ADL (the Architecture Definition Language)&lt;/strong&gt; and stored in &lt;code&gt;PromptModule&lt;/code&gt; CRs - never
as inline prose in an Agent spec. This guide covers how to write ADL and how modules
compose.&lt;/p&gt;
&lt;h2 id="why-adl"&gt;Why ADL&lt;/h2&gt;
&lt;p&gt;ADL expresses behavior as structured rules instead of free-flowing prose. The benefits
are practical: a rule-based prompt is &lt;strong&gt;scannable&lt;/strong&gt; (you can read what an agent will
and won&amp;rsquo;t do at a glance), &lt;strong&gt;versionable&lt;/strong&gt; (&lt;code&gt;kubectl diff&lt;/code&gt; shows exactly what changed),
and &lt;strong&gt;deployable without recompilation&lt;/strong&gt; (a &lt;code&gt;kubectl apply&lt;/code&gt; changes how an agent
reasons). Prose prompts drift and hide their intent; ADL keeps it explicit.&lt;/p&gt;</description></item><item><title>Define Fitness Functions</title><link>https://kubemoot.org/docs/user-guides/define-fitness-functions/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://kubemoot.org/docs/user-guides/define-fitness-functions/</guid><description>&lt;p&gt;A crew&amp;rsquo;s behavior is only as trustworthy as your ability to check it. Kubemoot
implements &lt;strong&gt;architectural fitness functions&lt;/strong&gt; as Kubernetes resources: you write a
short scenario that asks the crew a real question and states what a good answer must
satisfy, apply it, and the operator runs the crew live and scores each assertion pass
or fail.&lt;/p&gt;
&lt;h2 id="the-shape-of-a-scenario"&gt;The shape of a scenario&lt;/h2&gt;
&lt;p&gt;A fitness scenario is ADL pseudo-code - a question plus assertions:&lt;/p&gt;</description></item><item><title>Onboard MCP Tools</title><link>https://kubemoot.org/docs/user-guides/onboard-mcp-tools/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://kubemoot.org/docs/user-guides/onboard-mcp-tools/</guid><description>&lt;p&gt;A crew gains new abilities by gaining new &lt;strong&gt;tools&lt;/strong&gt;, and tools come from MCP servers.
You can declare them directly, or let Kubemoot&amp;rsquo;s &lt;strong&gt;autonomic onboarding&lt;/strong&gt; acquire them
when the crew hits a capability gap. This guide covers both.&lt;/p&gt;
&lt;h2 id="declare-a-tool-server-directly"&gt;Declare a tool server directly&lt;/h2&gt;
&lt;p&gt;When you already know the MCP server you want, declare an &lt;code&gt;MCPServer&lt;/code&gt; and let an
&lt;code&gt;MCPGateway&lt;/code&gt; discover it. The operator handles the deployment, health probes, and - for
stdio servers - the auto-injected &lt;code&gt;mcp-bridge&lt;/code&gt; sidecar.&lt;/p&gt;</description></item><item><title>kmctl User Guide</title><link>https://kubemoot.org/docs/user-guides/kmctl/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://kubemoot.org/docs/user-guides/kmctl/</guid><description>&lt;p&gt;&lt;code&gt;kmctl&lt;/code&gt; is the command-line tool for Kubemoot. This guide covers installation, a
runnable quickstart that walks through the core workflow, and shell-completion setup.&lt;/p&gt;
&lt;p&gt;For the complete command reference, see &lt;a href="../../reference/kmctl/"&gt;kmctl Reference&lt;/a&gt;.
For an overview of &lt;code&gt;kmctl&lt;/code&gt;&amp;rsquo;s role in the ecosystem, see
&lt;a href="../../ecosystem/kmctl/"&gt;kmctl - the CLI&lt;/a&gt;.&lt;/p&gt;
&lt;hr&gt;
&lt;h2 id="command-structure"&gt;Command structure&lt;/h2&gt;
&lt;p&gt;&lt;code&gt;kmctl&lt;/code&gt; uses a hybrid noun-verb layout, chosen deliberately as the best fit for a
domain CLI with deeply-nested resources and actions that have no clean kubectl
equivalent.&lt;/p&gt;</description></item></channel></rss>