openclaw

Orchestrate OpenClaw AI gateway workflows and agent routing through a Kubernetes control plane.

1|Updated Apr 3, 2026

Search Tags:#rbac #agent orchestration #mcp #kubernetes #telegram #openclaw #litellm

Ready to run

npx skills add https://github.com/villadalmine/infra --skill openclaw-villadalmine

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: openclaw
Source: https://github.com/villadalmine/infra/tree/main/skills/openclaw
Command: npx skills add https://github.com/villadalmine/infra --skill openclaw-villadalmine

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OpenClaw centralizes personal AI gateway traffic so messages, tool calls, and cluster actions flow through one controlled path instead of scattered direct API access. It is designed to keep model usage, RBAC, and network access auditable and safe in a Kubernetes homelab.

Core Features & Use Cases

Secure multi-agent orchestration across OpenClaw, Hermes, Kagent, and Kubernetes MCP tools.
LiteLLM-backed model routing with fallback chains, private API key handling, and model alias management.
Telegram, Honcho memory, webhooks, audio transcription, and operator troubleshooting for a production-style AI ops stack.
Use it to deploy the gateway, verify cluster visibility, and diagnose MCP or network policy issues from one skill.

Quick Start

Use the openclaw skill to deploy, configure, and troubleshoot the OpenClaw gateway for your cluster.

Frequently Asked Questions about openclaw

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I deploy a multi-agent AI gateway on Kubernetes?▼

Multi-agent AI gateway deployment on Kubernetes is orchestrated through a Kubernetes-native control plane that centralizes agent routing, applies readonly RBAC, and manages LiteLLM model routing with controlled egress.

What is the best way to centralize personal AI gateway traffic in a homelab?▼

Centralizing personal AI gateway traffic in a homelab is achieved by routing messages, tool calls, and cluster actions through a single controlled path, keeping model usage, RBAC, and network access auditable and safe.

Can I use LiteLLM routing with Telegram and MCP sidecars in Kubernetes?▼

LiteLLM routing works with Telegram integrations and MCP sidecars in Kubernetes by applying streamable-http MCP transport, managing private API keys, and handling persistent configuration across the cluster.

Do I need readonly RBAC to run multi-agent orchestration on a Kubernetes cluster?▼

Readonly RBAC is required for multi-agent orchestration on a Kubernetes cluster to ensure cluster inspection and operations remain auditable and safe within controlled egress network policies.

How do I troubleshoot MCP network policy issues in a Kubernetes AI gateway?▼

Troubleshooting MCP network policy issues in a Kubernetes AI gateway involves diagnosing MCP transport and network policies from a single orchestration skill, verifying cluster visibility and streamable-http MCP transport.

How does multi-agent delegation work for Kubernetes cluster inspection?▼

Multi-agent delegation for Kubernetes cluster inspection works by routing tool calls and cluster actions through OpenClaw, Hermes, Kagent, and Kubernetes MCP tools via a centralized control plane with readonly RBAC.