CubeSandbox gives every AI agent its own hardware-isolated MicroVM with a dedicated Linux kernel — not a shared-kernel container but a true KVM virtual machine — booting in under 60ms with less than 5MB of memory overhead per instance, enabling thousands of concurrent sandboxes on a single server via kernel sharing and Copy-on-Write memory. The Rust-built CubeProxy exposes an E2B SDK-compatible REST API on Axum, so existing E2B Cloud applications switch by changing one environment variable with zero client code changes. CubeMaster orchestrates cluster-level scheduling in Go while Cubelet manages node-local sandbox lifecycles including create, run, pause, resume, snapshot, and destroy operations. CubeVS, the eBPF-powered network layer, enforces per-sandbox isolation with LPM trie-based egress policies entirely in kernel space — no userspace context switches — while CubeEgress provides an L7 security proxy with per-domain, per-path, and per-method allowlists plus automatic credential injection that keeps secrets invisible to sandbox code. Hundred-millisecond-granularity snapshots enable checkpoint, rollback, and parallel fork from any saved state. The volume framework supports E2B-compatible pluggable storage with independent lifecycle and cross-sandbox sharing. Production deployment uses Terraform on Tencent Cloud or Helm charts on Kubernetes, K3s, or TKE, with full ARM64 support alongside x86_64 and AutoPause/AutoResume for cost optimization. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Agent code needs somewhere disposable to run, and Sandstorm turns one modest VPS into a fleet of thousands of isolated sandboxes that wake the moment work arrives. You create sandboxes through a plain HTTP API, starting from a registered template or forking a tagged snapshot, with per-sandbox memory ceilings, attached volumes, and egress rules set on each call. Every sandbox is a gVisor container with its own network namespace, so untrusted agent code runs away from your host and away from its neighbors. Idle sandboxes cool down through four tiers, from running hot, to frozen in place with a wake time near ten milliseconds, to checkpointed on disk and back in about a fifth of a second, so a small machine can hold far more sandboxes than it could ever keep running at once. A single binary acts as master, node, and web console: the master tracks desired state and places work across nodes, nodes behind NAT dial out to join the fleet, and a gateway on each node wakes a sleeping sandbox and forwards its traffic automatically. The built-in Hub pulls ready-made snapshot packs for warmed runtimes like Python with numpy or a pre-warmed Playwright browser, skipping cold starts entirely. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

Describe an app in one sentence and sandboxd, a self-hosted AI app builder, has a coding agent build it in an isolated sandbox and hand back a live preview URL. Start from a blank prompt, a React, Next.js, Express, or FastAPI starter, or an imported public Git repo. Refine it in one persistent chat beside Preview, Code, Terminal, and Git tabs. Pick OpenCode, Pi, or Claude Code. Point OpenCode or Pi at local or remote OpenAI-compatible and Anthropic-compatible endpoints, LiteLLM gateways included. A proxy injects provider keys on the wire, so nothing secret lands in a sandbox. Did the agent get it wrong? Every task is checkpointed first, so one click reverts the turn. Snapshots, forks, and named versions keep known-good points to branch from. Edit files in the browser, review file-by-file diffs, commit, and push. Secrets are write-only and encrypted per app. Idle sandboxes sleep and wake on the next request. One modest server can host per-user previews, coding playgrounds, or an agent platform. The app library installs 80+ curated open-source apps, such as Ghost, n8n, Gitea, and Grafana, each in its own sandbox. Every console action is also a /v1 HTTP call with an OpenAPI spec. Upgrades back up the database, health-check the new version, and roll back on failure. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

OpenAgentCore puts Codex, Claude Code, and MiniMax Code behind one self-hosted Agents API, so your applications start agent sessions through the official OpenAI SDK while the work runs on machines you control. Set OPENAI_BASE_URL to your installation and issue a Project API key, and the same client code that talks to OpenAI now talks to your Core. Each session runs a native harness, exactly as it runs on a developer's machine, with the model provider, model ID, and protocol you configure per harness or per session. Choose where sessions execute: microsandbox nodes you operate, Docker, E2B, or a self-hosted workstation, VM, or container connected by the oac-daemon with a credential scoped to one environment. Idle microsandbox sessions release their compute and resume on a compatible node with workspace files intact, so long investigations do not rent a machine while they sleep. The Web console walks operators through setup, tracks sessions and turns, manages models and credentials, and issues keys, while the separate Core API serves operators and the Agents API serves applications. Durable execution state in PostgreSQL keeps events, results, and schedules recoverable across restarts. Sandbox providers, workspace filesystems, harnesses, and model providers all plug in through defined protocols, so every layer can be replaced on its own. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Eclipse Che spins up complete development environments on Kubernetes from a Devfile in the project repository, so new developers click a factory link and open a browser IDE with dependencies, runtimes, and debugging already configured instead of spending their first day installing toolchains. Each workspace runs as a DevWorkspace custom resource in the user's namespace, making environments API-manageable with kubectl and GitOps tooling alike. Developers choose their editor per workspace, from Visual Studio Code Open Source to JetBrains IDEs via Projector, and the 7.124 release adds Orca, an agent development environment for working alongside AI coding agents. Workspaces carry their own containers for build and run, so code executes in a production-like environment with direct Kubernetes API access from the IDE. Teams standardize stacks as Devfile templates, onboard contributors in minutes, and open a fresh workspace per branch or pull request in a new browser tab. The 7.124 release also builds workspace lifecycle commands into chectl, including SSH connections with port forwarding to desktop editors, and DevWorkspace Operator 1.0 adds optional per-workspace NetworkPolicy isolation. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. EPL-2.0 licensed.

With over 14,000 GitHub stars and enterprise adoption by security-conscious organizations, Coder transforms how development teams provision, manage, and secure their coding environments. Every workspace is defined as a Terraform template, meaning infrastructure engineers can standardize development environments across EC2 instances, Kubernetes pods, Docker containers, or any combination, while developers get self-service provisioning that launches in seconds rather than days of manual setup. The WireGuard-based networking layer establishes encrypted tunnels between developer machines and remote workspaces, providing low-latency access without exposing ports or configuring VPN concentrators. Automatic idle detection shuts down unused workspaces after configurable periods, directly reducing cloud compute costs for organizations running hundreds of developer environments. The Coder Agents feature introduces native AI coding capabilities where the agent loop executes entirely within the control plane on self-hosted infrastructure, keeping LLM API credentials out of individual workspaces and eliminating credential exfiltration risks. Centralized model governance allows platform teams to approve specific AI providers and models, set per-user spend limits, and maintain complete audit logs of all prompts, tool calls, and agent activity. IDE integration supports VS Code through a dedicated extension, JetBrains IDEs via Gateway and Toolbox plugins, and browser-based code-server for web access. The template registry provides pre-built configurations for common development stacks. DevContainer support builds environments from standard devcontainer.json specifications. Deploy on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

AgentBox is a self-hosted browser workspace that runs Claude Code and Codex CLI on a server you control, so a task goes from chat to reviewed code without leaving one place. Open a workspace from an empty project, an uploaded archive, or a Git clone, then talk to the agent in streaming web chat or drop into its original interactive terminal when you want the CLI itself. Files persist between sessions along with your home configuration and chat history, so a project picked up on a laptop continues on a phone exactly where it stopped. Review per-file diffs line by line before committing, pushing, or opening a pull request, and download the result when no Git connection is configured. Shared files, layered home templates, Claude skills, and MCP configuration carry your setup into every new workspace, with per-workspace overrides when a project differs. A trusted team shares central account pools with per-user scopes, while every model request is recorded with tokens, cache hit rate, cost, and latency, exportable to CSV and optionally capped by credits. An optional tunnel reaches allowlisted repositories and databases on private networks, and six interface styles with light and dark modes keep long sessions comfortable. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

Created by the developer behind Uptime Kuma and carrying over 23,000 GitHub stars, Dockge brings the same clean, reactive design philosophy to Docker Compose stack management with a web interface that makes Portainer's compose handling feel like editing YAML in a terminal over SSH. The interactive editor provides syntax highlighting, inline validation, and a live preview of your compose.yaml files while keeping every stack stored as a standard file on disk in /opt/stacks by default, meaning you can seamlessly switch between the web UI and the docker compose CLI without lock-in or proprietary database formats. Real-time WebSocket updates stream pull progress, container start/stop transitions, and build output directly to the browser with no polling delays. The built-in web terminal opens a shell session inside any running container for quick debugging, while the docker-run-to-compose converter transforms single-container run commands into proper compose.yaml definitions with one click. Multi-agent support introduced in version 1.4.0 connects multiple Docker hosts to a single Dockge dashboard, enabling centralized management of stacks distributed across different servers. Image update detection shows which stacks have newer versions available, and one-click updates pull the latest images and recreate containers without manual intervention. The stack is a single Docker container running on Node.js with Socket.IO for reactivity and stores no external database. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

For self-hosters who want Git-driven container deployments without surrendering control of their existing reverse proxy or host networking, Dockerizalo provides a lightweight web platform that clones repositories, builds Docker images, and manages running services from a single dashboard. Point an application entry at any Git-compatible repository or public container registry image, configure custom build paths, and let the platform build and launch the container directly on your host Docker engine. A built-in template library accelerates database provisioning with pre-configured variables, networks, and persistent storage mounts. Through the web interface, operators manage environment secrets, bind-mounted volumes, custom Docker labels, and published TCP or UDP ports while coexisting peacefully alongside Caddy, Nginx, or Traefik setups. Per-application HTTP webhook triggers let teams automate continuous delivery from GitHub, GitLab, Gitea, or custom CI pipelines on every push event. Real-time paginated build and runtime container log streams prevent memory exhaustion on chatty APIs while keeping every deployment event visible. Configurable image retention policies automatically prune older tags and remove dangling images after each rollout so disk space stays clean. Multi-user account management allows homelab administrators to invite collaborators to the dashboard without sharing a single root credential. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Komodo puts every server, container, and deployment pipeline behind a single dashboard where you build, ship, and monitor Docker workloads across unlimited hosts. Lightweight stateless agents install on each connected server and report CPU, memory, and disk metrics back to the Rust-powered core, giving you real-time visibility without heavyweight monitoring stacks. Docker Compose stacks deploy directly from the UI or link to Git repositories with webhook-triggered automatic redeploys on push. A built-in CI pipeline compiles versioned Docker images from source, with optional AWS spot instances for burst build capacity. For orchestration at scale, Docker Swarm management handles node configuration, service deployment, and multi-node stack orchestration from the same control plane. Browser terminal sessions open persistent named shells on servers and inside containers, complete with shared team access and scriptable Actions that chain executions into multi-stage procedures. Infrastructure-as-code support defines all resources declaratively in TOML files within a Git repository, keeping production state version-controlled and auditable. Granular role-based access control with user groups, per-resource permissions, and OAuth through GitHub and Google keeps teams operating within defined boundaries. A full OpenAPI specification, dedicated CLI, and typesafe client libraries for Rust and TypeScript make programmatic integration straightforward. With 12,000+ stars and active development, the community continues expanding multi-architecture builds and Swarm tooling.

Scatter a few cheap VPS boxes, a home server, and one bare metal machine across three different providers, and Uncloud still treats them as a single place to run your applications. You describe services in ordinary Docker Compose files, deploy them with the uc command line tool, and publish any container port as a public HTTPS endpoint in a single command. Every machine joins an encrypted WireGuard mesh with its own container IP addresses, so services on different continents reach each other directly without VPN configuration or exposed ports. A built-in DNS server resolves service names to container IPs, and the integrated Caddy reverse proxy provisions and renews Let's Encrypt certificates automatically, sharing them across the cluster. Rolling updates replace containers with zero downtime, persistent Docker volumes keep stateful services safe, and the Unregistry integration pushes image layers straight to your machines when no registry is available. There is no control plane to babysit: each machine keeps a synchronised copy of the cluster state, and SSH access to any single node is enough to manage the whole fleet. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

One server, treated properly: Nanocl runs your containers and full virtual machines from a single declarative Statefile, reconciling what is running against what the file says, so a lone host gets cluster discipline without cluster ceremony. Containers group into Cargoes with replica counts, ordered init containers, and health-aware restarts; finite jobs run to completion and stop. When a workload needs a real kernel or a custom disk image, it boots as a QEMU virtual machine beside the containers instead of forcing a second platform. Publishing is built in: the proxy controller terminates TLS and routes HTTP and TCP traffic, and the DNS controller keeps service names resolving while workloads come and go. Secrets for registries, certificates, and environment variables attach where they are used, never inside the file itself. The daily loop is a CLI and an API: list processes, follow logs, exec into a container, watch events, and back up the entire deployment state before you change it. Namespaces keep projects apart on the same host, and Statefile arguments, templates, and sub-states carry one definition from staging to production. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT or Apache-2.0 licensed.

ToolHive is an open-source MCP server management platform that lets you run, secure, and orchestrate Model Context Protocol servers in isolated containers, giving AI agents structured access to tools like GitHub, databases, and cloud services without exposing host credentials or network. Every server launches inside its own sandboxed container with a minimal permission file, network access filtering, and encrypted secrets management, preventing misbehaving connectors from reaching beyond their defined scope. The built-in registry provides a catalog of vetted servers you can install with one command, while custom images and package-manager references let you onboard proprietary connectors without writing Dockerfiles. Platform teams deploy the Kubernetes operator to declare MCP servers as cluster resources using Custom Resource Definitions, with automated lifecycle management and multi-namespace isolation. The Virtual MCP Server gateway aggregates multiple backends behind a single endpoint, centralizing OIDC authentication, tool filtering, and composite cross-server workflows so clients connect once instead of juggling separate URLs. An MCP Optimizer analyzes tool schemas via semantic search and surfaces only relevant tools per request, cutting token consumption by up to 85%. OpenTelemetry traces and Prometheus metrics deliver full visibility into tool execution, latency, and request audit trails across every managed server. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

A single GPU can only hold so many models at once, and when a chat server, an image pipeline, and a voice engine all grab VRAM together, somebody runs out of memory and crashes. Berth fixes the scramble by making every engine lease the card: wrap a launch in berth run and the daemon grants memory in turn, queues the rest with an honest place in line and a start estimate, and drains a holder cleanly before the next load begins. The broker learns each engine's real footprint from its runs, so grants are sized by measured peaks instead of guesses, and processes that grab memory without a lease are reported as stowaways, never killed, with contested grants flagged in plain sight. A collisions board counts what the scheduling prevented: avoided collisions, VRAM handed off, drains, stuck releases, and refusals, and every count links back to a gap free event log you can follow live. The Harbor web UI turns that into a departures board of leases, a six hour tide of memory by holder, engine cards with decode throughput and a token ledger, plus a per model leaderboard and thirty days of stored history. Alerts reach your desktop or a webhook, Prometheus scrapes the metrics endpoint, and multi GPU machines can even pass one card through to a VM while the others keep serving. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

Your inference server should not sit mostly idle while a model crawls along at ten tokens per second, so ReadyLLM gives you a visual way to deploy, watch, and tune LLMs until the hardware is actually earning its keep. Add the machine under your desk or a remote box over SSH, let hardware detection read its GPU, VRAM, CPU, and memory, and get model recommendations that fit what you own instead of wishful downloads. A model store catalogs text, image, and video models with parameter counts, precision, resolutions, and real weight filenames, background downloads from multiple sources, and integrity checks when they land. One click starts or stops llama.cpp, vLLM, or SGLang services, installs a missing engine matched to your GPU backend, and leaves runtime parameters open for manual tweaks. Live curves track GPU utilization, memory, throughput, and latency as requests flow. When speed disappoints, two-phase auto tuning searches the knobs that matter, checks VRAM feasibility before it wastes your time, and benchmarks honestly against latency, throughput, or prefill goals; an AI tuning mode can reason through the same loop across several rounds. Beyond chat models, ComfyUI generation covers text-to-image and text-to-video, storyboards, shot-by-shot long video, and upscaling, with prompts optionally expanded first. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.