The ConnectWise Automate MCP Server - free, local, built for MSPs
Independent, open source, inspectable. Every line of code is on GitHub under Apache-2.0 - built for the MSP community, vendor-neutral by design. Not affiliated with, endorsed by, or sponsored by ConnectWise, LLC.
Passes all 4 mechanical gates (build · command-surface · claims · install). Awaiting its first MSP receipt - be the first, 60 seconds →.
Yes - there is an MCP server for ConnectWise Automate. It’s free, open source, and runs on your own machine, so your client data never leaves your network. It connects ConnectWise Automate to Claude, ChatGPT, Copilot, or any MCP-capable agent, and installs in about 60 seconds.
Ask your AI “which clients are behind on patches?” or “where are my offline agents?” and get a fleet-wide answer in one shot. ConnectWise Automate’s console is per-server and built for one endpoint at a time. This skill syncs your whole fleet - computers, clients, alerts, patch history - into a local SQLite mirror, then answers the cross-client questions the web UI can’t.
New to the term? An MCP server is the same thing ChatGPT calls an app or connector, Claude on the web calls a connector, and Claude Code calls a Skill. One thing, many names →
Install in 60s → View on GitHub →
Instead of clicking through ConnectWise Automate, just ask
Instead of Logging into each client’s Automate view and tallying offline agents by hand before a license true-up
just ask: “Which agents haven’t checked in for 30+ days, grouped by client?”
Your agent runs: connectwise-automate-cli stale-agents --days 30 --agent
Instead of Pulling a patch report per client in Patch Manager and reassembling the numbers for a QBR
just ask: “Which clients are behind on patches, worst first?”
Your agent runs: connectwise-automate-cli patch-compliance --agent
Instead of Clicking through the Monitors and Alerts views server by server to see what needs a tech this morning
just ask: “What open alerts need a human across every client right now?”
Your agent runs: connectwise-automate-cli alert-triage --min-priority 3 --agent
See it in 30 seconds
Demo data is simulated. Every command shown exists in the real CLI.
What it does
| Question your MSP keeps asking | Command your agent runs |
|---|---|
| Which agents haven’t checked in for 30+ days, by client? | connectwise-automate-cli stale-agents --days 30 --agent |
| Which clients are behind on patches, worst first? | connectwise-automate-cli patch-compliance --agent |
| What open alerts need a human across every client? | connectwise-automate-cli alert-triage --min-priority 3 --agent |
| What’s my whole-fleet health right now - online/offline, last contact, open alerts? | connectwise-automate-cli fleet-health --agent |
| Give me a one-line snapshot per client for the review. | connectwise-automate-cli client-rollup --agent |
| What operating systems are end-of-life across the fleet? | connectwise-automate-cli os-inventory --eol-only --agent |
| What changed overnight - new alerts, check-ins, patches installed? | connectwise-automate-cli since --hours 24 --agent |
Full command reference at github.com/servosity/msp-skills/blob/main/skills/connectwise-automate/guide.md.
What makes this one different
Most ConnectWise Automate integrations proxy each question into a live API call - fine for one computer, useless when you ask “across all 40 clients.” This skill syncs your fleet into a local SQLite mirror, so a cross-client roll-up is one local join: instant, offline, and your AI sees the answer, not thousands of raw agent records.
Automate’s console and reports are per-server and built for drilling into one endpoint. This complements them by answering the whole-book-of-business questions - stale agents, patch posture, EOL exposure, overnight drift - in a single command, without replacing your RMM.
The pain this closes
- Automate’s console is per-server and one-endpoint-at-a-time - there’s no native cross-client roll-up, so a simple “how many offline agents do I have across all clients?” means logging into views and tallying by hand.
- Offline and stale agents quietly inflate your license count and hide risk, and finding them before a true-up or security review is manual.
- Patch compliance for a QBR lives in Automate, but pulling it per client and reassembling it into something you can hand a client eats an afternoon.
Install
Works in any of these agents - pick yours:
| Agent | Quick install |
|---|---|
| Claude Desktop | Step-by-step → |
| ChatGPT (Plus/Pro+) | Step-by-step → |
| Claude Code | Step-by-step → |
| Codex CLI | Step-by-step → |
| Cursor, Windsurf, Cline, Continue, Zed, Copilot, Gemini, Hermes, OpenClaw | Which agent? → |
Quickest path for everyone else (terminal):
macOS / Linux:
bash <(curl -fsSL https://raw.githubusercontent.com/servosity/msp-skills/main/skills/connectwise-automate/install.sh)
Windows (PowerShell):
iwr -useb https://raw.githubusercontent.com/servosity/msp-skills/main/skills/connectwise-automate/install.ps1 | iex
After install, authenticate once with your ConnectWise Automate credentials, then verify with connectwise-automate-cli --version.
Safety model
| Tier | Examples | Recommended agent policy |
|---|---|---|
| Read | fleet-health, stale-agents, patch-compliance, alert-triage, client-rollup, os-inventory, since, and every list / get / search command | Allow |
| Endpoint and fleet actions | computers command-execute (runs a real command on an agent), patching deploy-approved / deploy-security / reattempt-failed (fleet-wide patch deployment) | Human-in-the-loop, explicit confirmation |
| Write (data) | import |
Preview with –dry-run, then a reviewed write |
| Credential | apitoken mint, apitoken refresh, auth set-token | Human-in-the-loop only |
The skill is read-first: fleet roll-ups, stale-agent sweeps, patch posture, alert triage, inventory, and search are all read-only and safe to let an agent run. A small set of commands change real endpoints - running a command on an agent (computers command-execute) and deploying patches across the fleet (patching deploy-approved / deploy-security / reattempt-failed) - plus token minting; keep those human-in-the-loop. The credential’s own ConnectWise Automate permissions are the outer limit on anything the agent can do. Full details in governance.md.
Frequently asked questions
Is there an MCP server for ConnectWise Automate?
Yes - this one. A free, open source MCP server and Claude Code Skill for ConnectWise Automate, built for MSPs. It runs locally on your machine, works with Claude, ChatGPT, Copilot, and any MCP-capable agent, and installs in about 60 seconds.
Is the ConnectWise Automate MCP server safe for client data?
Yes, by design. The CLI, the MCP server, and any local data mirror run on your own machine - nothing is sent to MSP Skills or any third party. Credentials stay in your environment, and every command is safety-tiered (read, write, destructive) so your agent only gets the permissions you grant. Full policy in the safety model on this page.
Does this work with ChatGPT?
Yes, on paid ChatGPT plans. ChatGPT connects to remote MCP servers over HTTPS, so you expose the local ConnectWise Automate MCP server via a secure bridge. Step-by-step in the install guide.
Do I need to know how to code?
No. Paste one sentence into Claude Code or Codex and your agent does the install, or run a one-line installer. You enter your credentials once.
Is my ConnectWise Automate data safe?
Your data stays on your machine. The CLI, MCP server, and the local mirror are all local. The AI sees query results, not raw bulk data, and credentials are never bundled or transmitted by MSP Skills.
What does it cost?
Free. Apache-2.0 licensed. You pay only for whichever AI agent you already use.
Is this for ConnectWise Automate or ConnectWise Manage (PSA)?
ConnectWise Automate - the RMM, formerly LabTech. ConnectWise Manage (the PSA) is a separate skill. This one talks to your Automate server’s API for computers, clients, alerts, patching, scripts, and monitors.
Will this hit my Automate API rate limits?
The local mirror is the point. You sync once, then every roll-up, triage, and search runs against local SQLite, so day-to-day questions never touch the API. Only sync and the live list/get commands call Automate.
Do I need to be a ConnectWise partner?
You need an Automate server you administer plus an API token, and a registered integration clientId GUID (required for v2020.11+). The skill authenticates as you and adds nothing to your ConnectWise account.
More RMM connectors
Run more than one RMM tool, or comparing options? These connectors work the same way: Action1 · Atera · Datto RMM · Level · N-able N-central · Nerdio Manager · NinjaOne · Tactical RMM
Status
Beta. Validated against the ConnectWise Automate API surface and being validated with MSPs running it live against their own production tenants in our weekly Build Sessions.
Build Sessions are free and stay free - The Build Room is where the deep work happens.
Standards. Conforms to the open Agent Skills spec (Anthropic, Dec 2025; 40+ agents). MCP-compatible - works with any MCP-capable agent including Hermes. OpenClaw-ready (frontmatter pre-wired, awaiting OpenClaw launch).
Maintained by Servosity for the MSP community. Apache-2.0 licensed. Built with CLI Printing Press.