
ConnectWise Manage MCP
io.github.ServosityPublished by GitHub account Servosityv0.2.4Updated Oct 24, 2026
Every ConnectWise PSA workflow from the terminal - with a typed conditions query builder, offline
Overview
Local MCP server that gives an AI assistant access to ConnectWise PSA (Manage) data and workflows, with a local SQLite mirror and cross-entity views.
- What you can do
- Query cross-entity views: unbilled tickets, agreement burn, board triage, stale tickets, technician workload, and a single-client account card
- Search and read records from a local SQLite mirror with full-text search, offline after an initial sync
- Build validated ConnectWise conditions filter expressions for API queries
- Run typed subcommands across the PSA REST surface (service, time, company, finance, project, sales, procurement, system), each also exposed as an MCP tool
- Create and update records such as time entries, tickets, and companies, and delete tickets, companies, or configurations
- When to use it
- Useful for MSP owners and technicians who want to ask cross-entity questions of their ConnectWise PSA from an AI agent, such as which tickets have no time logged or which agreements are near their block-hour limit, without exporting to spreadsheets or clicking through portal screens.
- Requirements
- Runs locally as a stdio process, installed from an .mcpb bundle or an installer script; the MCP server shells out to a companion CLI binary. Needs a ConnectWise API Member public and private key, a client ID from the ConnectWise developer portal, and the company ID and region or on-prem host, supplied through CW_COMPANY_ID, CW_PUBLIC_KEY, CW_PRIVATE_KEY, CW_CLIENT_ID and CW_SITE. An initial sync is needed before cross-table questions.
Installation
In SourceWeft
- Open ConnectWise Manage MCP in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
ConnectWise Manage + AI - for Claude, the ChatGPT desktop app, GitHub Copilot, Gemini CLI, and any agent that speaks MCP
Unofficial. Community-built Claude Code Skill and MCP server for the ConnectWise Manage API. Not affiliated with, endorsed by, or sponsored by ConnectWise, LLC.
[ConnectWise PSA (Manage) demo - animated preview]
▶ Watch the 30-second demo - demo data is simulated; every command shown exists in the real CLI.
Every ConnectWise PSA workflow from the terminal - with a typed conditions query builder, offline SQLite sync, and cross-entity views (unbilled work, account 360, board triage) the PSA web UI can't give you. Works with the AI you already use - ChatGPT (desktop app), Claude Desktop, Codex, Claude Code, Claude Cowork, and GitHub Copilot - plus Gemini CLI. Free, open source, runs on your laptop. Built for MSP owners. No code required.
Works with your agent
Security note (until this connector is reprinted). Do not run HTTP mode (
--transport http), not even on localhost. This build has no password, listens on every network interface, and predates the browser-rebinding guard, so a web page you open can reach it. The MCP tools can also make changes with your credentials. Use the default stdio transport. If you set up a tunnel for a remote agent, shut it down. A fix is on the way.
The six agents MSP owners actually use (self-serve, works today):
The ChatGPT desktop app runs connectwise-manage-mcp locally and shares MCP configuration with Codex (OpenAI). ChatGPT on the web needs a remote (HTTP) server, and this connector's HTTP mode is paused until it is reprinted (see the security note in mcp-install.md), so msp-skills does not support it yet. If your team works there, a hosted gateway is the better pick; check that your AI app is on the gateway's setup list.
Also for the Microsoft and Google stacks
Big install base, but an honest heads-up: Gemini CLI runs the local binary you just installed, while Microsoft 365 Copilot and the Gemini app need a remote server, which msp-skills does not support yet.
Skill-native agents (also covered): Hermes and OpenClaw read this skill's
SKILL.mddirectly and speak MCP - see their install sections below. Also works with Cursor, Windsurf, Cline, Continue.dev, and Zed via MCP. Full per-tool wire-up: docs/which-agent.md.
Run more than one agent? Install across all 51+ supported agents in one command:
npx skills add Servosity/msp-skills@latest(requires Node.js, then run the per-skill installer for the CLI/MCP binaries). See docs/which-agent.md.
Install in 60 seconds
Fastest for Claude Desktop - one-click .mcpb
Claude Desktop bundle: the
.mcpblaunches on macOS (Intel and Apple Silicon), Windows x64 and Linux x64. Every tool shells out to the companionconnectwise-manage-cli. Releases cut from 2026-09-18 on ship that CLI inside the bundle; older bundles contain only the MCP server, so if yours lacks the companion, run the Path A or Path B installer below first (or setCONNECTWISE_MANAGE_CLI_PATHto an existing binary). Details: #331.
Download ConnectWise Manage MCP (.mcpb) - then open Claude Desktop > Settings > Extensions and select the file. One click, no JSON, no shell. (Browse every ConnectWise Manage release on the releases page.)
Prefer the Claude Code plugin? Add the marketplace once, then install - works immediately, no directory listing required:
Path A - paste one prompt into your AI agent (recommended)
Copy this into Claude Code, Codex CLI, or Claude Cowork:
Install the ConnectWise Manage Skill and MCP server from Servosity/msp-skills in this agent workspace. If this workspace uses a POSIX shell (macOS, Linux, WSL, or Bash), run
bash <(curl -fsSL https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/connectwise-manage/install.sh). If it uses Windows PowerShell, runiwr -useb https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/connectwise-manage/install.ps1 | iex. Then authenticate per the README and runconnectwise-manage-cli --helpto explore.
The same prompt works in any agent that can run shell.
Path B - run the installer yourself
Windows (PowerShell):
macOS / Linux:
The installer drops both connectwise-manage-cli (the CLI) and connectwise-manage-mcp (the MCP server) into your user bin path. Claude Code, Codex, and Cowork discover the Skill via SKILL.md in this directory.
Verify:
Upgrade to the latest version
The installer always fetches the current release - re-run it to upgrade:
macOS / Linux:
Windows (PowerShell):
Claude Desktop .mcpb users: download the latest .mcpb (top of this section) and re-select it in Settings > Extensions. Claude Code plugin users: /plugin update connectwise-manage@msp-skills.
Add to Claude Desktop, the ChatGPT desktop app, GitHub Copilot, Gemini CLI, or another MCP client
After the installer runs, see mcp-install.md and docs/which-agent.md for the per-agent wire-up - one section per agent, including the GitHub Copilot servers key and the ChatGPT desktop app. Claude Desktop's Settings > Extensions panel is the simplest path; the MCP config block (for users who prefer editing JSON) is documented in mcp-install.md.
Install for Hermes
From the Hermes CLI:
Inside a Hermes chat session:
Hermes speaks MCP natively, so it can also use the connectwise-manage-mcp server directly - same install path, same env vars.
Install for OpenClaw
Tell your OpenClaw agent (copy this):
Install the connectwise-manage skill from https://github.com/servosity/msp-skills/tree/main/skills/connectwise-manage. The skill defines how its required CLI (
connectwise-manage-cli) can be installed via theopenclaw:frontmatter block.
OpenClaw isn't generally available yet; the frontmatter wiring is pre-shipped and will activate the moment OpenClaw launches.
Authenticate
Set the credentials the CLI needs - an API Member with public/private keys from your Manage instance (System > Members > API Members) plus a clientId from the ConnectWise developer portal:
doctor confirms the credentials work before you run anything that touches data. It exits 0 even when the credential is rejected, so scripts should add --fail-on error. The API Member's security role is the permission boundary - scope it to what you want the AI to reach. Reading the product catalog (procurement get-catalog, sync --resources procurement-catalog) needs Procurement > Product Catalog inquire rights on that role, or ConnectWise refuses the catalog calls.
What this skill does
Beyond the views: the full ConnectWise PSA REST surface (service, time, company, finance, project, sales, procurement, system) is exposed as typed subcommands - service get-tickets, time post-entries, company get-companies and hundreds more - every one also available as an MCP tool.
Full command reference: guide.md. For the AI-agent operating contract (--agent, --dry-run, when to confirm before mutating), see AGENTS.md.
What makes this different
Most ConnectWise Manage integrations and MCP servers proxy each question into a live API call. That's fine for one record. It dies at scale, when you're asking "which tickets did we close across every client this month with no time logged against them" - and every live call has to thread the strict conditions syntax, where one wrong quote returns silently empty results.
This skill syncs ConnectWise Manage into a local SQLite mirror with full-text search. Aggregate questions become one local SQL join: instant, offline, and the AI sees the answer, not the raw data. Compound commands like unbilled, account, and agreement-burn join across tickets, time entries, companies, contacts, agreements, and configurations - work a stateless API wrapper can't do. And when you do need a live filtered query, condition build assembles a validated conditions expression so the API call is right the first time.
The pain this closes
Unlogged time is direct revenue leakage. MSP consultancy Bering McKinley prices the cost plainly - "lost billable hours, lower profitability, bad decision-making caused by incomplete data, and unhappy clients" - and third-party reviews call time entry the most consistently reported ConnectWise pain point in the MSP community. Techs close tickets without entering time, and nobody notices until the invoice run, because Manage has no single view that joins closed tickets to their logged hours. The same export-to-Excel reflex shows up for agreement burn and client-360 questions: the data is in the PSA, but composing it across entities means portal screen after portal screen or a paid BI add-on.
What this skill does about it:
connectwise-manage-cli unbilled --since 7d- tickets you touched or closed this week with zero (or under-threshold) hours logged. Run it before every billing cutoff.connectwise-manage-cli agreement-burn --period 30d- hours consumed vs allotment per agreement, with an over-limit flag. Spot the unprofitable client before they blow the block.connectwise-manage-cli stale --days 5- open tickets rotting with no update, oldest first. The daily board-hygiene pass.connectwise-manage-cli workload- open count and oldest age per tech, so the next ticket goes to whoever is lightest.connectwise-manage-cli account AcmeCorp- the five-screen client picture in one card.
See pain-point.md for the longer narrative.
Frequently asked questions
Is there an MCP server for ConnectWise Manage?
Yes. This is a free, open source MCP server for ConnectWise Manage (now called ConnectWise PSA). It runs on your own machine, works with cloud and on-premises Manage, and connects to Claude Desktop, Claude Code, and other MCP-capable agents. Every install path is in Install in 60 seconds, and the agent setup guide wires it into any other agent. Run connectwise-manage-cli sync once (or ask your agent to sync ConnectWise) before your first cross-table question.
Other community MCP servers for ConnectWise Manage include WYRE's connectwise-manage-mcp and MSPStack's mcp-connectwise-psa. This one keeps a local SQLite copy of your PSA, so questions that join tickets, time, and agreements are answered from the local copy without calling the API again, on cloud or on-premises Manage. MSPStack's server can also answer cross-table questions with read-only SQL against the Manage database, which its README documents for on-premises installs; this one needs only API access, not a database login. This one also has a condition build command that writes valid conditions filters.
Does this work with ChatGPT?
Yes, in the ChatGPT desktop app. It runs MCP servers locally (STDIO), the way connectwise-manage-mcp runs, and shares MCP configuration with Codex (OpenAI): add connectwise-manage-mcp under Settings > MCP servers as a STDIO server, or add it once to Codex's ~/.codex/config.toml. ChatGPT on the web is not supported by msp-skills yet: it needs a remote (HTTP) server, and this connector's HTTP mode is paused until it is reprinted; the security note in mcp-install.md explains why. If your team works there, a hosted gateway is the better pick; check that your AI app is on the gateway's setup list.
Does this work with Codex, Cursor, Windsurf, Cline, Copilot, or Gemini?
Yes - all of them speak MCP. Cross-tool install commands are in the matrix above and the deep-dive in docs/which-agent.md.
Do I need to know how to code?
No. The recommended install is to paste one sentence into Claude Code or Codex - your agent reads SKILL.md and does the install. The fallback is a one-line installer per OS (bash or PowerShell). Neither path requires writing code. You'll enter your ConnectWise Manage credentials once.
Is my ConnectWise Manage data safe?
Your data stays on your machine. The CLI and MCP server are local binaries. The SQLite mirror sits in a directory under your user account. The AI agent only sees what the CLI returns - typically a query result, not raw bulk data. Credentials are read from your environment or your agent's config; never bundled into this repo or transmitted anywhere by MSP Skills.
Will this hit my ConnectWise API rate limits?
The local mirror exists so reads stop hitting the API. After the first sync, the cross-entity views (unbilled, account, agreement-burn, board, stale, workload) run against local SQLite with zero API calls. Live calls respect a --rate-limit throttle. Note: ConnectWise's list endpoints expose no "changed since" parameter, so a plain sync re-fetches each resource in full rather than only what changed (you'll see resource_not_incremental warnings) - on a tenant with years of history, scope the sync with a conditions filter as shown below.
I have years of history - how do I keep sync and reads fast?
ConnectWise list endpoints declare no temporal-filter parameter, so a plain sync re-fetches every resource in full and --since has no effect on the sync itself. Two levers keep a heavy tenant fast:
- Scope the sync with a conditions filter.
synctakes no resource positional: name the resource with--resources, and attach the condition with--resource-param <resource>:<key>=<value>(repeat the pair per resource you want to bound). Build a valid expression withcondition buildfirst: This fetches only the slice you ask for instead of the whole history. - Stay local for reads. The cross-entity views read the local mirror.
searchdefaults to auto mode (it tries the live API first); add--data-source local(and optionally--type <resource>) to stay fully offline, and give a slow command more room with--timeout 2m. The MCPsearchtool is already local-only, so agents on the MCP server never hit the live-search path.
Is this for ConnectWise PSA or ConnectWise Manage?
Same product - ConnectWise renamed Manage to ConnectWise PSA. This skill targets the Manage REST API (v4_6_release/apis/3.0), which is the API both names refer to. It does not cover ConnectWise Automate - that's a separate product with a separate API.
Does it work with on-premises ConnectWise Manage?
Yes. CW_SITE accepts your own server's hostname as well as the cloud region hosts; the CLI builds the standard v4_6_release/apis/3.0 base URL either way.
Will this replace my ConnectWise portal?
No - and it isn't trying to. The portal stays best for in-app workflows like drag-and-drop dispatch and invoice editing. This skill brings the PSA's data to whichever AI agent you already use, and answers the cross-entity questions no single portal screen composes.
What does it cost?
Free. Apache-2.0 licensed. You pay only for whichever AI agent you use (Claude, ChatGPT, Codex, etc.), and that's billed by your AI provider, not by us.
Safety model
Writes are not gated in the binary by default - --dry-run previews a request without sending, but it's a flag your agent must pass. Put the gate in your agent's policy: preview, show the exact command, get approval, then run.
The strongest control is the scope you grant the ConnectWise Manage credentials - the CLI can only do what the credentials are permitted to do. Full details, including how to lock it down, are in governance.md.
Status
Beta. Validated against the ConnectWise Manage API surface and being validated with MSPs running it live against their own production tenant in our weekly Build Sessions. RSVP at compoundingteams.com/build-sessions.
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. Apache-2.0 licensed. Built with CLI Printing Press. Last updated: 2026-06-05.
Source: skills/connectwise-manage/README.md at commit 061ec05
Tools
0Version history
6- v0.2.4LatestOct 23, 2026
- v0.2.3Oct 23, 2026
- v0.2.2Oct 22, 2026
- v0.2.1Oct 16, 2026
- v0.2.0Sep 29, 2026
- v0.1.8Sep 16, 2026
Score
40% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Declare a license in the repository or the registry entry.