About DataGate MCP
An open-source CLI and MCP server for MSPs and telecom resellers that bill through DataGate. It provides read-only lookups for customers, agreements, invoices, and other resources, syncing data to a local SQLite mirror to reduce repeated API calls.
Key features
- 01Look up customers, users, and agreements
- 02Retrieve invoices and service items
- 03Cover multiple DataGate resource groups
- 04Sync data to a local SQLite mirror
- 05Search the local mirror with full-text search
Use cases
- 01Review agreements and service items by customer
- 02Find invoices within a date range
- 03Repeat customer and billing record lookups
- 04Search resources in the local mirror
MCP support
Provides an MCP server for compatible AI clients to query data through the DataGate API. The connector is read-only and uses a local SQLite mirror for repeated lookups.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEDataGate + AI - for Claude, ChatGPT, Codex, Cursor, and any agent that speaks MCP
Unofficial. Community-built Claude Code Skill and MCP server for the DataGate telecom billing API. Not affiliated with, endorsed by, or sponsored by DataGate. DataGate is a trademark of its respective owner.
Add DataGate customer, agreement, and invoice lookups to the AI you already use - Claude Code, Claude Desktop, ChatGPT (Plus/Pro+, paused until this connector is reprinted; see mcp-install.md), Codex, Cursor, Windsurf, Cline, Continue, Gemini, or GitHub Copilot. Free, open source, runs on your laptop. A local SQLite mirror means repeated lookups don't have to page through the live API by hand, and don't eat into DataGate's per-account rate limit. Built for MSPs and telecom resellers who bill through DataGate. 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 checks a token, but it predates the fix release's request size and time limits. 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.
| Your AI agent | How to install the DataGate skill |
|---|---|
| Claude Desktop | Run installer, then Settings > Extensions to register datagate-mcp (no JSON editing). |
| ChatGPT (paid plans) | Paused until this connector is reprinted: ChatGPT needs its HTTP mode, and in this build that mode checks a token but has no request size or time limits yet. See the security note in mcp-install.md. |
| Claude Code | Paste the install prompt below. |
| Codex CLI | Paste the install prompt below. |
ChatGPT needs a remote HTTPS server, which means this connector's HTTP mode, and that is paused until the connector is reprinted: in this build that mode checks a token but has no request size or time limits yet. See the security note in mcp-install.md.
Also works with Cursor, Windsurf, Cline, Continue.dev, Zed, GitHub Copilot, and Gemini CLI - plus any MCP-speaking agent, via MCP and the pre-wired skill frontmatter. Full per-tool wire-up: docs/which-agent.md.
Install
macOS / Linux:
bash <(curl -fsSL https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/datagate/install.sh)
Windows (PowerShell):
iwr -useb https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/datagate/install.ps1 | iex
Both installers detect your OS and architecture, download the datagate-cli and
datagate-mcp binaries from this repo's Releases, write them to ~/.local/bin
(macOS/Linux) or %LOCALAPPDATA%\Programs\msp-skills\ (Windows), clear macOS
Gatekeeper quarantine, and honor DRY_RUN=1.
Authenticate
export DATAGATE_API_KEY="<your Bearer token>"
export DATAGATE_CLIENT_ID="<your ClientId GUID>"
datagate-cli doctor
Both values come from your own DataGate account - there's no OAuth flow, no token
expiry, and nothing to refresh. doctor confirms both are set and the API is
reachable.
Why a local mirror instead of a wrapper
Most thin DataGate wrappers proxy every question into a live API call. That's fine for one customer lookup, but it adds up once you're checking a large customer list or pulling invoices repeatedly - each call counts against DataGate's rate limit (60 requests/minute, 5,000/day, per account). This skill syncs customers, agreements, and invoices into a local SQLite mirror with full-text search, so repeated lookups become one local query: instant, offline, and no extra API calls.
What it covers
Read-only (list, get, search) across DataGate's resource groups: customers, customer users, agreements, service items, assignments, rate cards, sites, product templates, kit templates, invoices, products, delivery methods, product transactions, and account managers. See governance.md for the full read/write breakdown and guide.md for every command.
Example commands
datagate-cli customers list --json
datagate-cli agreements list --select id,customerId,name,status --json
datagate-cli invoices --period-start 2026-08-01T00:00:00Z --period-end 2026-08-31T23:59:59Z --json
datagate-cli sync
datagate-cli search "<customer name or invoice number>"
Known gaps
This build is read-only by design - DataGate's CRUD surface is documented but this connector doesn't implement create/update/delete. A handful of endpoint paths (sites, products, product templates, kit templates, delivery methods, product transactions, account managers) were derived from DataGate's own naming convention rather than individually confirmed against a live tenant - if you run this against your own account and something's off, please file it (see CONTRIBUTING.md, rung 2 - a bug report with real-tenant evidence is worth more than one we can't reproduce).
License
Apache-2.0. See cli/LICENSE and cli/NOTICE.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.