OpenClaw Helm Chart

# OpenClaw Helm Chart

[![Chart Version](https://img.shields.io/badge/chart-v0.0.0--pre--alpha-orange.svg)](https://github.com/5dlabs/openclaw-helm/releases)
[![License: AGPL-3.0](https://img.shields.io/badge/License-AGPL%203.0-blue.svg)](LICENSE)
[![GitHub Stars](https://img.shields.io/github/stars/5dlabs/openclaw-helm?style=social)](https://github.com/5dlabs/openclaw-helm)
[![Discord](https://img.shields.io/discord/1234567890?label=Discord&logo=discord)](https://discord.gg/openclaw)
[![Build Status](https://github.com/5dlabs/openclaw-helm/workflows/Lint%20and%20Test/badge.svg)](https://github.com/5dlabs/openclaw-helm/actions)

> **⚠️ EARLY DEVELOPMENT**: This chart is under active development and not yet ready for production use. The API and configuration format may change. Check back in a few days for the stable release.

A Kubernetes Helm chart for deploying OpenClaw AI agents with comprehensive messaging platform support, observability, SSH remote development, and inter-agent communication.

## About

[OpenClaw](https://github.com/openclaw/openclaw) is an open-source personal AI assistant that runs on your own devices, answering you on the channels you already use. Created by the OpenClaw project, it supports 14+ messaging platforms with a local-first Gateway control plane.

**This Helm chart** provides Kubernetes deployment automation for OpenClaw, created and maintained by 5D Labs. The chart packages OpenClaw with enterprise-grade features including multi-replica support, observability, SSH remote development, and production secrets management.

**CTO Platform**: This chart is a subset of the [CTO platform](https://cto.5dlabs.ai/) — a complete autonomous engineering team with 13 AI agents, desktop app interface, and 60-80% cost savings vs cloud. [Learn more →](CTO-PLATFORM.md)

**Monorepo Structure**: This repository includes TypeScript packages and an interactive installer. For development and package documentation, see [MONOREPO.md](MONOREPO.md).

### What's Included in the Container

Each agent container includes:

- **7 AI Coding CLIs** (all run headless): Claude Code, Cursor, Codex, Every Code, Factory, Gemini, OpenCode
- **Claude Code with Team Support**: Official Anthropic CLI with full team collaboration for multi-agent workflows
- **6 Language Runtimes**: Node.js 22 LTS, Python 3, Go, Rust, PHP, Ruby
- **Observability Stack**: Grafana + Loki for comprehensive logs, metrics, and dashboards
- **High-Performance Code Editing**: Morph MCP server (10,500 tok/s fast editing and intelligent search)
- **Browser Automation**: Playwright + agent-browser CLI for headless web automation
- **Security Suite**: Gitleaks, Trivy, Hadolint, Semgrep for comprehensive code scanning
- **1Password Integration**: Headless secret access via `op` CLI with service account tokens
- **React Native Development**: EAS CLI and Expo for mobile agent development
- **Database Clients**: PostgreSQL, MySQL, Redis, SQLite pre-installed
- **Kubernetes Tools**: kubectl, helm, argocd, k9s, stern, kustomize
- **SSH Server**: Remote development support (VS Code Remote, JetBrains Gateway)
- **9 Messaging Platforms**: Discord, Slack, Telegram, WhatsApp, Signal, Matrix, Teams, Google Chat, WebChat
- **Skills System**: Agent Skills spec (agentskills.io) compliant with dual security scanning
- **Inter-agent Communication**: NATS-based messaging for multi-agent collaboration

## Features

- **Claude Code Team Support**: Full multi-agent collaboration with shared workspaces and inter-agent messaging
- **Multi-platform Messaging**: Discord, Slack, Telegram, WhatsApp, Signal, Matrix, Teams, Google Chat, WebChat
- **Multiple AI Model Providers**: Anthropic Claude, OpenAI, OpenRouter
- **Built-in Observability**: Grafana + Loki for comprehensive monitoring
- **NATS Inter-Agent Messaging**: Pub/sub communication for coordinated multi-agent workflows
- **SSH Remote Development**: Full VS Code Remote and JetBrains Gateway support
- **WireGuard VPN**: Secure connectivity for distributed agents
- **Flexible Secrets Management**: Inline or external Kubernetes Secrets
- **Extensible Skills System**: Agent Skills spec compliant with dual security scanning
- **Guided TUI Install**: Interactive terminal UI for non-infrastructure users
- **Horizontal Scaling**: Multi-replica deployments with Git worktree isolation
- **Production-Ready Resource Management**: Configurable limits, persistence, RBAC

## Quick Start with KinD

```bash
# Create a local Kubernetes cluster
kind create cluster --name openclaw-demo

# Create secrets first (v0.2.0+ requires external K8s Secrets)
kubectl create secret generic openclaw-api-keys \
  --from-literal=ANTHROPIC_API_KEY=sk-ant-xxx

kubectl create secret generic openclaw-messaging \
  --from-literal=DISCORD_TOKEN=your-token

kubectl create secret generic openclaw-github \
  --from-literal=GITHUB_PAT=ghp-xxx

# Install the chart
helm install my-agent oci://ghcr.io/5dlabs/charts/openclaw \
  -f agents/demo/values.yaml

# Chec

... (truncated)