kaalm
Deploy, manage, and secure AI agents on Kubernetes using the Kaalm operator.
Free
Works with the AI Agents you already use
kaalm
Example session with this skill installed
I want to deploy a long-lived agent to my local k3d-kaalm-tutorial cluster to handle support tickets. I have an OpenAI API key ready.
- Read your context and instructions
- Compiled the kaalm
I will help you deploy a support agent to the k3d-kaalm-tutorial cluster. First, we will set up the ModelProvider for OpenAI, then create the Agent resource.
On context k3d-kaalm-tutorial, this creates a Secret for your API key and a ModelProvider for the agent to use.
Agent Deployment Manifest
apiVersion: kaalm.io/v1alpha1
kind: ModelProvider
metadata:
name: openai-provider
namespace: kaalm-agents
spec:
type: openai
secretRef:
name: openai-api-key
---
apiVersion: kaalm.io/v1alpha1
kind: Agent
metadata:
name: support-agent
namespace: kaalm-agents
spec:
agentClass: standard-llm
modelProviderRef:
name: openai-provider
image: ghcr.io/win07xp/kaalm-sdk-python:latest
env:
- name: AGENT_ROLE
value: "support"
Next steps
- Run
kubectl --context k3d-kaalm-tutorial create secret generic openai-api-key --from-literal=api-key=YOUR_KEY_HERE -n kaalm-agentsto securely add your credential. - Run
kubectl --context k3d-kaalm-tutorial apply -f agent-spec.yaml --server-side --dry-run=serverto verify the manifests. - Once the dry run passes, I will apply the configuration to start the support agent pod.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing AI agents on Kubernetes often forces developers to choose between complex manual YAML orchestration or opaque black-box platforms. Scaling agent workloads, managing model budgets, and handling secret rotation across a cluster adds significant operational overhead. This skill provides a structured interface for Kaalm, a Kubernetes operator designed specifically for agentic resources. It automates the deployment and lifecycle management of agents while maintaining strict security boundaries within your existing infrastructure.
What it does
- Resource Management deploys long-lived agents or ephemeral tasks using native Kubernetes objects.
- Budget & Access Control configures model providers, agent classes, and spending limits for multi-tenant team environments.
- Security Orchestration handles API keys and secrets without ever exposing raw values to the LLM or logs.
- Tool Integration enables agents to interact with external tools and channels through defined gateway protocols.
- Cluster Diagnostics identifies and resolves issues with resources that fail to reach a Ready state.
How it works
- Context Initialization reads local configuration to verify the target cluster, user experience level, and deployment goals.
- Resource Definition generates or modifies manifests for Agents, AgentTasks, and ModelProviders based on provided examples.
- Safety Validation performs server-side dry runs on all cluster changes before requesting user confirmation to apply.
- Lifecycle Execution manages the installation, upgrading, and monitoring of the Kaalm operator via Helm and kubectl.
Frameworks & tools
This skill requires kubectl and helm installed on the local machine. It interacts with Kubernetes clusters and is optimized for Kaalm v2.1.0 resources, including specialized CRDs for agents and model gateways.
Why this beats prompting it yourself
Writing Kubernetes manifests for specialized operators is error-prone and requires constant reference to technical documentation. This skill enforces safety patterns like mandatory context checking and dry runs that prevent accidental cluster misconfiguration. It automatically adapts its explanations and technical depth based on whether you are a platform engineer or new to Kubernetes.
Use cases
- Deploying Agent Fleets to handle persistent background processing on a production cluster.
- Setting Up Team Environments with shared model providers and strict usage budgets.
- Automating Agent Tasks that run as completion-based jobs for data processing or audits.
- Securing LLM Access by routing model calls through a controlled gateway with managed API keys.
Known limitations
Does not support reading or printing raw Secret values or API keys by design. Requires a local Python environment to manage persistent configuration files.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.
Cannot read or print raw Secret values. Requires local Python for config management. Specific to Kaalm v2.1.0 CRDs and CLI arguments.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 19 days ago
- Free with an Agensi account
- Accounts
GitHub
- Software
Python
GitkubectlHelm
Permissions
Allowed Hosts