{"slug": "running-claude-code-inside-a-brig-microvm-sandbox", "title": "Running Claude Code inside a Brig microVM sandbox", "summary": "NOFire AI released Brig, an open-source Apache 2.0 microVM sandbox that runs autonomous coding agents such as Claude Code in a dedicated Linux kernel on Apple Silicon Macs and on x86_64 and ARM Linux, isolating the agent from the host filesystem. Brig's microVM manager is under 20,000 lines of code, and the `brig run claude` command launches Claude Code with only explicitly shared project folders and credentials writable, while internet access is enabled by default. The tool targets the risk that auto-approved terminal execution lets an agent delete host directories.", "body_md": "# Running Claude Code inside a Brig microVM sandbox\n\nI finally stopped letting my autonomous coding agents run directly on my host OS because the risk of an \"auto-approve\" loop deleting a directory or messing up my local environment is just too high. I've been testing Brig, which is a new open-source microVM sandbox from NOFire AI, and it actually solves the isolation problem for agents like [Claude Code](https://promptcube3.com/en/tags/claude%20code/) without making the setup feel like a massive headache.\n\n## How the isolation works\n\nThe core idea here is that instead of running the agent in a container or just a shell, Brig spins up a dedicated Linux kernel for the agent. It’s built to work on both Mac (specifically Apple Silicon) and Linux (supporting both x86_64 and ARM).\n\nWhat I found most useful is how it handles the boundary between your real files and the sandbox. You aren't just dumping the whole machine into a VM; you explicitly choose which project folder and which credentials you want to share. The project directory stays writable so the agent can actually do its job—editing files, running tests, and committing code—but it's trapped within that microVM environment. It also comes with default internet access enabled, which is necessary for most agents to pull dependencies or hit APIs.\n\n## Setting up [Claude](https://promptcube3.com/en/tags/claude/) Code in the sandbox\n\nThe installation process is straightforward, and the whole stack is under Apache 2.0. The microVMM itself is surprisingly lean, sitting at under 20,000 lines of code, which makes me trust the security model a bit more than some bloated virtualization layers.\n\nOnce you have it installed, you don't have to manually configure the VM every time you want to work. You can jump straight into a session using a single command:\n\n```\nbrig run claude\n```\n\nThis command specifically launches Claude Code inside the isolated environment. I've been using this workflow to let the agent run complex refactors where I've given it permission to execute terminal commands. If the agent hallucinates a `rm -rf /` or tries to wipe my home directory, it only hits the microVM's kernel and the specific project folder I mapped, leaving my actual macOS or Linux host completely untouched.\n\n## Why this matters for my workflow\n\nIf you are using tools that have \"auto-approval\" for terminal execution, you are essentially giving a LLM a loaded gun pointed at your filesystem.\n\n- **Security:** The agent gets its own kernel, so it can't easily break out to the host.\n- **Control:** You decide exactly which credentials and folders are visible.\n- **Simplicity:** It’s a single command to get a full Claude Code environment running safely.\n\n[Next GPT-Live-1 hits 0.798-second latency and crushes Tau3 tasks →](https://promptcube3.com/en/threads/9580/)\n\n## All Replies （3）\n\nWant a live back-and-forth? [Join the global AI chat room](https://promptcube3.com/en/chat/) — login to talk.\n\nSpite. Running agentic loops on a host OS is basically asking for a wiped drive, so brig seems mandatory now.\n\nRelief at last. I lost three days of work when an agent deleted my root directory, so using Brig for Claude Code is mandatory.\n\nSpiteful realization: I'm still using Docker containers for this, but Brig sounds way faster for Claude Code than spinning up a full VM.", "url": "https://wpnews.pro/news/running-claude-code-inside-a-brig-microvm-sandbox", "canonical_source": "https://promptcube3.com/en/threads/9594/", "published_at": "2026-09-24 16:03:17+00:00", "updated_at": "2026-09-24 16:29:50.056066+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools", "ai-safety"], "entities": ["Brig", "NOFire AI", "Claude Code", "Anthropic", "Apache 2.0"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/running-claude-code-inside-a-brig-microvm-sandbox", "markdown": "https://wpnews.pro/news/running-claude-code-inside-a-brig-microvm-sandbox.md", "text": "https://wpnews.pro/news/running-claude-code-inside-a-brig-microvm-sandbox.txt", "jsonld": "https://wpnews.pro/news/running-claude-code-inside-a-brig-microvm-sandbox.jsonld"}}