coder/coder
Secure environments for developers and their agents
What it solves
Coder provides a self-hosted platform to manage cloud development environments (CDEs) and AI coding agents. It eliminates the long onboarding process for developers by providing pre-configured workspaces that can be run on any infrastructure (EC2, Kubernetes, Docker), while reducing costs by automatically shutting down idle resources.
How it works
Workspaces are defined using Terraform templates, allowing infrastructure to be managed as code. These environments are connected via secure Wireguard tunnels. For AI capabilities, Coder Agents runs native AI coding agents on the user's own infrastructure, utilizing an AI Gateway to centralize authentication, auditing, and cost controls without requiring LLM credentials to be stored within the individual workspaces.
Who it’s for
Software developers and platform engineers who want to maintain full control over their development environments and self-host their AI coding tools on their own infrastructure for security and governance.
Highlights
- Infrastructure as Code: Uses Terraform to define environments on EC2, Kubernetes, or Docker.
- ** Coinside with AI Agents**: Supports native AI coding agents with support for various models (Anthropic, OpenAI, Google, Bedrock, self-hosted).
- Cost Efficiency: Automatically shuts down idle workspaces to save on cloud spend.
- Secure Access: Uses Wireguard tunnels for secure connectivity to workspaces.
- Centralized AI Governance: Provides an AI Gateway for model governance, cost tracking, and audit logging.
Related
- Project
newsnowlabs/docksideA self-hosted platform for launching isolated, browser-accessible development containers for every branch or AI agent session on your own infrastructure.
- Project
tastyeffectco/sandboxdA self-hosted AI app builder: one HTTP request spawns an isolated container, runs a coding agent against your prompt, and returns a live preview URL — with your own models, data, and infrastructure.
The idea: instead of sending your code to a hosted builder like Bolt or v0, you own the whole stack. sandboxd is intentionally small — Go + Docker + Traefik + SQLite, no Kubernetes — and favors operational simplicity: idle containers sleep and wake automatically, so many apps fit on one machine.
Two ways to use it: a REST API for embedding into products, or a no-code web console (a pure API client) for interactive building. Beyond AI generation, it can launch 80+ curated open-source apps as one-click containers.
Not for everyone: the README says skip it if you just need a couple of containers (a Docker script suffices). It's aimed at scenarios like AI app-builder products, agent platforms, per-user preview environments, or team multi-app hosting. Monetization: MIT-licensed core, optional managed cloud and BYOC services.
- Project
Mng-dev-ai/agentroveAgentrove is a self-hosted, open-source (Apache 2.0) AI coding workspace that unifies seven coding agents into one web app, gives each project its own isolated workspace with chat/editor/terminal/git, and lets agents orchestrate each other via sub-threads and git worktrees — plus group "channels" where you and multiple agents collaborate in one shared folder. It runs as a Docker web app, a macOS Tauri app, or a thin iOS client for your own server.
- Project
Abilityai/trinityTrinity is an open‑source, self‑hosted platform that runs persistent AI agents as if they were employees. Agents live in Docker containers, have Git‑backed state, and are governed by roles, objectives, and gated approvals recorded in an immutable audit log. The system supports Claude Code, OpenAI Codex, and Gemini CLI back‑ends, provides a Vue‑based Workspace UI, and offers enterprise‑grade security and compliance extensions.
- Project
openchamber/openchamberOpenChamber is an open‑source, cross‑platform workspace that lets developers run AI coding agents, set automated goals, compare multiple model runs, review changes with guided walkthroughs, and ship code—all from desktop, web, VS Code, or mobile.