Stackdome: Open Source Self-Hostable Railway Alternative on Kubernetes

Stackdome is an open source, self-hostable deployment platform that wraps the CNCF ecosystem into an opinionated, Railway-style experience. It runs on Kubernetes but hides the complexity, letting developers deploy apps without diving into YAMLology. The project started in 2024 as a hobby and is now self-hostable from a small VPS to a large cluster.
Key Features
- Multi-service canvas: A visual UI (inspired by Railway) shows your entire stack. You can edit, wire up, and create resources directly on the canvas.
- Multi-cluster support: Stackdome acts as a hub, placing workloads across multiple Kubernetes clusters in a hub-and-spoke pattern.
- Managed Postgres: Built on CNPG, it includes WAL archiving, high availability, backups, and point-in-time recovery (PITR).
- Release system: Deploy a release that targets the whole stack and roll it back as a unit if something breaks.
- Source-to-image: Uses an in-cluster registry (Zot) to build and deploy images directly from source.
- Preview environments: Disposable, multi-service preview environments for testing pull requests.
- Declarative apps: Define your stack in a
Stackfileand deploy via the CLI.
Getting Started
To install on a VPS, run:
curl -fsSL https://get.stackdome.com/install.sh | sudo sh
You can also try the hosted cloud at cloud.stackdome.com – no credit card or compute required.
Who It's For
If you already run Kubernetes and want a higher-level developer experience for your team, Stackdome is something you can adopt incrementally. It's also a strong option if you want a self-hosted Platform-as-a-Service that doesn't lock you into a vendor.
At the time of writing, the project is on GitHub with early adoption. The developer mentions ideas for an 'expert mode' to expose more Kubernetes bells and whistles, plus the ability to adopt already-running workloads – so the roadmap is worth watching.
📖 Read the full source: HN LLM Tools
👀 See Also

Claude Code protocol file reduces repetitive questioning
A developer created a single .md file for ~/.claude/rules/ that infers task type and risk from the first message, eliminating Claude Code's typical three-question sequence before starting work.

Two MCP Tools for Claude Code: Idea Validation and Trading Agent Memory
A developer built two MCP tools for Claude Code: idea-reality-mcp checks GitHub and Hacker News before coding to avoid duplicates, while tradememory-protocol provides memory for AI trading agents to store trades with context and track strategy performance. Both are open source and available on PyPI.

Essential OpenClaw plugins for developers using AI coding agents
A developer tested OpenClaw plugins and identified essential tools including env-guard for security, commit-guard for preventing bad commits, composio for connecting to 860+ tools, cortex-memory for long sessions, cost-tracker for spending visibility, and openclaw-better-gateway for fixing flaky connections.

Claude File History: VS Code Extension for Tracking Claude Code Sessions
A VS Code extension called Claude File History tracks every Claude Code session that touched your files, allowing you to find past conversations, preview what was discussed, and resume conversations with a double-click.