ghl-mcp-proxy
A minimal multi-sub-account MCP proxy for GoHighLevel that allows listing configured accounts and making authenticated REST API calls on behalf of any account.
README
ghl-mcp-proxy
A minimal multi-sub-account MCP proxy for GoHighLevel / LeadConnector. Holds a Private Integration Token (PIT) + Location ID per sub-account and exposes two tools to Claude:
list_accounts— shows which sub-accounts are configuredghl_api_request— makes an authenticated call to the LeadConnector REST API on behalf of a chosen account
1. Configure your accounts
Copy .env.example to .env and fill in:
PROXY_SHARED_SECRET— a long random string (openssl rand -hex 32). This is the "password" Cowork will send back to this server.GHL_ACCOUNTS_JSON— one entry per sub-account, keyed by a friendly name:
{
"acme_dental": { "pit": "pit-xxxxxxxx", "locationId": "loc-xxxxxxxx" },
"joes_plumbing": { "pit": "pit-yyyyyyyy", "locationId": "loc-yyyyyyyy" }
}
Keep this as valid JSON on one line in the real .env file.
2. Deploy on Hostinger
Option A — Hostinger VPS (recommended, most control)
-
SSH into your VPS.
-
Install Node 20+ and git if not already present:
curl -fsSL https://deb.nodesource.com/setup_20.x | sudo bash - sudo apt-get install -y nodejs git -
Upload this project (via
git clonefrom your own repo, orscp). -
cd ghl-mcp-proxy && cp .env.example .envthen edit.envwith real values. -
Install dependencies and run with a process manager so it survives reboots:
npm install --omit=dev npm install -g pm2 pm2 start src/server.js --name ghl-mcp-proxy pm2 save pm2 startup # follow the printed instructions -
Put Nginx in front of it for HTTPS (Hostinger VPS panel can issue a free Let's Encrypt cert). Point a subdomain, e.g.
mcp.yourdomain.com, at the VPS, then reverse-proxy port 443 →localhost:8080.Minimal Nginx server block:
server { listen 443 ssl; server_name mcp.yourdomain.com; ssl_certificate /etc/letsencrypt/live/mcp.yourdomain.com/fullchain.pem; ssl_certificate_key /etc/letsencrypt/live/mcp.yourdomain.com/privkey.pem; location / { proxy_pass http://localhost:8080; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; } } -
Test:
curl https://mcp.yourdomain.com/healthshould return{"ok":true,"accounts":2}.
Option B — Hostinger Node.js App (hPanel, if on a plan that supports it)
- In hPanel, create a new Node.js application, point it at this project (upload via File Manager or Git).
- Set the entry point to
src/server.js. - Add all variables from
.env.exampleunder the app's Environment Variables section (do not upload a real.envfile to a shared host). - Set the app's port to match what Hostinger assigns via
process.env.PORT(already handled inserver.js). - Attach your domain/subdomain and enable SSL (Hostinger issues this automatically for most plans).
- Restart the app, then test
https://your-subdomain/health.
3. Add it as a custom connector in Cowork
- In Claude/Cowork: Customize → Connectors → "+".
- Name:
GoHighLevel Proxy(or anything). - URL:
https://mcp.yourdomain.com/mcp - Advanced settings → set a custom header:
Authorization: Bearer <your PROXY_SHARED_SECRET> - Click Add, then Connect.
4. Use it
Ask Claude something like:
"List my connected GHL accounts, then for acme_dental find contacts tagged 'hot lead' and summarize what's happening with them."
Claude will call list_accounts, then ghl_api_request with
account: "acme_dental" and the right REST path.
Adding more sub-accounts later
Just add another entry to GHL_ACCOUNTS_JSON and restart the app
(pm2 restart ghl-mcp-proxy on VPS, or restart via hPanel). No code changes
needed.
Security notes
- This single server holds credentials for all your clients — treat the
shared secret and
.envfile like a master password. - Use the narrowest PIT scopes each workflow actually needs.
- Rotate PITs periodically from GHL Settings → Private Integrations.
- Never commit
.envto git (already covered by.gitignore).
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.