clawops is a CLI and MCP server for deploying and operating self-hosted OpenClaw instances on AWS, GCP, Azure, or Linux VMs. It provides typed MCP tools and safety controls for day-to-day infrastructure operations like status, logs, SSH, backups, upgrades, and stack management.
From the registry: Deploy and manage self-hosted OpenClaw instances across clouds
$ npm install -g @clawops/clihttps://www.npmjs.com/package/@clawops/cli$ which clawopsPlease install the `clawops` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Node.js 22 or newer** — Node.js version 22 or higher is required to install and run the clawops CLI/MCP server.
- **Install the clawops CLI** — Install the published CLI package globally so the MCP client can launch the clawops binary. Run: `npm install -g @clawops/cli` (https://www.npmjs.com/package/@clawops/cli)
- **Determine the absolute path to the clawops binary** — Use the absolute binary path in MCP configs so GUI clients can launch clawops reliably. Run: `which clawops`
Optional prerequisites:
- Configure cloud credentials if you will use cloud providers — clawops does not store cloud credentials and reads them from your environment or provider CLI login. AWS uses AWS_PROFILE or the standard AWS credential chain, GCP uses GOOGLE_APPLICATION_CREDENTIALS or application-default login, and Azure uses AZURE_CLIENT_ID/AZURE_CLIENT_SECRET or az login.
- Have SSH access set up for local/existing-host deployments — For local or existing server management, you need an SSH key and an SSH-reachable Linux or macOS host.
Canonical MCP server config (stdio transport):
- command: `<absolute-path-to-clawops>`
- args: ["mcp","serve","--read-only"]
Note: README recommends starting with --read-only for first evaluation. Full stdio mode uses args ["mcp","serve"]. The server also supports HTTP mode via `clawops mcp serve --http <port> --bind <host>`, but the README does not provide a canonical MCP client HTTP config snippet, so stdio is the preferred extracted config.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Real-time Kubernetes network traffic visibility and API analysis for HTTP, gRPC, Redis, Kafka, DNS.
Gives AI development tools Firebase-specific capabilities and expertise.
All Azure MCP tools to create a seamless connection between AI agents and Azure services.
MCP tools for interacting with Microsoft Fabric