thinkjones

AI Openclaw Lmstudio — DevOps skill for Claude Code

DevOps community

Run OpenClaw AI coding agent in a Docker Sandbox with LM Studio as the local LLM backend.

How to install AI Openclaw Lmstudio

This entry records only its repository, not the path inside it, so there is no exact command to give. Open thinkjones/ai-openclaw-lmstudio and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What AI Openclaw Lmstudio does

Run OpenClaw AI coding agent in a Docker Sandbox with LM Studio as the local LLM backend.

Alternatives in DevOps

Skill Claw — OpenClaw instance administration — manage hosts across macOS, Ubuntu/Debian, Docker, OCI, and Proxmox 2.8k ★
Claudebox — A Claude Code Docker Development Environment for running Claude AI's coding assistant in a fully containerized 964 ★
Codeg — Aggregate and browse AI coding agent sessions (Claude Code, Codex, Gemini CLI, etc.) in one place 850 ★

README

openclaw-lmstudio

Run [OpenClaw](https://docs.openclaw.ai) (AI coding agent) inside a Docker Sandbox with your choice of LLM backend:

LM Studio — Local inference, free, full privacy
Claude — Anthropic API, best-in-class reasoning

Inspired by [Run OpenClaw Securely in Docker Sandboxes](https://www.docker.com/blog/run-openclaw-securely-in-docker-sandboxes/).

Prerequisites

Docker Desktop (macOS/Windows) or Docker Engine + Compose v2 (Linux)
LM Studio (if using local models) — installed with a model downloaded and server running
Anthropic API key (if using Claude) — from console.anthropic.com

Quick Start

There are two ways to get started — choose the one that fits your situation:

Path A: Fresh Install (no existing OpenClaw)

Use this if you've never installed OpenClaw before, or want to start from a clean config.

# 1. Clone the repo
git clone https://github.com/thinkjones/openclaw-lmstudio.git
cd openclaw-lmstudio

# 2. Configure
cp .env.example .env
# Edit .env — choose your provider and fill in the required variables

# 3. Run
chmod +x scripts/setup.sh scripts/start.sh
./scripts/setup.sh

`setup.sh` generates a fresh `openclaw.json` from your `.env`, creates the `.openclaw-files/` directory structure, builds the Docker image, and starts the container.

Path B: Migrate Existing Config (recommended if you have `~/.openclaw`)

Use this if you already have OpenClaw installed at `~/.openclaw` with auth profiles, API keys, and settings configured.

# 1. Clone the repo
git clone https://github.com/thinkjones/openclaw-lmstudio.git
cd openclaw-lmstudio

# 2. Copy your existing config
chmod +x scripts/copy-config.sh scripts/start.sh
./scripts/copy-config.sh

# 3. Build and run
docker compose build
docker compose up -d

`copy-config.sh` copies your auth profiles, credentials, agent configs, and settings into `.openclaw-files/.openclaw/`. It patches `gatew