Openclaw Monorepo — DevOps skill for Claude Code
Repo-local OpenClaw workspace with modular JSON5 config, plugins, and Docker sandboxes — runs coding-agent CLIs on desktop or Android Termux.
How to install Openclaw Monorepo
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open appautomaton/openclaw-monorepo and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Openclaw Monorepo does
Repo-local OpenClaw workspace with modular JSON5 config, plugins, and Docker sandboxes — runs coding-agent CLIs on desktop or Android Termux.
Alternatives in DevOps
- Skill Claw — OpenClaw instance administration — manage hosts across macOS, Ubuntu/Debian, Docker, OCI, and Proxmox 2.8k ★
- Codeg — Aggregate and browse AI coding agent sessions (Claude Code, Codex, Gemini CLI, etc.) in one place 850 ★
- Workers MCP — Talk to a Cloudflare Worker from Claude Desktop 646 ★
README
openclaw-platform
Monorepo-style **OpenClaw** workspace for local usage + development (git submodule in `openclaw/`) with modular JSON5 config and optional Docker sandboxes.
- English (this file) · 中文: `README.zh-CN.md`
- Run everything via
pnpm openclaw ...(wrapper: `scripts/openclaw.mjs`) - Commit-safe config lives in `config/openclaw/` (split into modular JSON5 files)
- Repo-local state lives in
bots/(gitignored; migration notes in `bots/README.md`) - Instead of the default
~/.openclaw/, this repo pins state/config/workspaces underbots/(e.g.bots/workspaces//) so everything stays in one place
[!NOTE] You are reading the `monorepo` branch. If you need the Termux + proot setup and tooling, switch to the `proot-debian` branch.
[!IMPORTANT] `config/.env` and `bots/` contain secrets/state. Keep them out of git.
This repo intentionally tracks only `bots/README.md` and `bots/openclaw.json` under `bots/`.
What this repo gives you
- A predictable, repo-local OpenClaw setup: config + workspaces + state live together under this repo folder (instead of
~/.openclaw/). - Modular JSON5 config you can edit/review in git (secrets stay in
config/.env, never in JSON5). - Repo-local plugins (custom tools/skills) under
plugins/that live outside theopenclaw/submodule. - Optional Docker sandbox images for agents that need extra system dependencies (LibreOffice, OCR, etc.).
Contents
- Prerequisites
- Quick start
- How config is wired
- Repo layout
- Project tree (sketch)
- Docs
- Common commands
- Sandboxed agents (Docker)
Prerequisites
- Node.js 24.x (recommended; OpenClaw requires
>=22.12.0) corepack enable(forpnpm)- Docker (optional; only needed for sandboxed agen
Related Skills
Infrastructure Sandbox Kit
Docker Sandboxes template + kit: an IaC (Pulumi/Terraform/OpenTofu + AWS/Azure/GCP CLIs) Claude Code workstati
AI Openclaw Lmstudio
Run OpenClaw AI coding agent in a Docker Sandbox with LM Studio as the local LLM backend
Openclaw Obsidian Desktop
Docker setup for running OpenClaw and Obsidian together in a browser-accessible virtual desktop, with an OpenC
Agentswarm
Local agent swarm: a conductor model splits missions into parallel sub-agents with real tools and a localhost
Besvelte
A Svelte implementation of bots in the app. Claude, Codex, Opencode, others - create bots, run jobs across vir
LiteEditorAI
Terminal-first workspace for supervising AI coding agents (Claude Code, Codex, Gemini): live agent status per
Related Agents
Openclaw Debugger
Use when an OpenClaw instance has a problem the health sentinel couldn't fix — gateway failures, config drift,
Workspace Cartographer
Maps the territory of this monorepo. Answers structural questions like "where does X live?", "which project ow
Docker Deploy
Use to build and ship container images: multi-stage builds, transitive monorepo dependencies, digest pinning,