Setup
1. Install dependencies
bun install
2. Create and place serviceAccount.json
To allow the server to access Firebase Remote Config, you need a Google Cloud service account key file:
A. Google Cloud Console (point‑and‑click)
- Open IAM & Admin → Service Accounts inside the same GCP project that owns your Firebase app.
- Click Create Service Account
- Name:
mcp-remote-config
(any name is fine)
- Description: “MCP server – Remote Config access”
- Grant this service account access:
- In the role picker, search for Remote Config Viewer or Remote Config Admin (as needed) and select it.
- Optionally add Firebase Analytics Viewer if your template conditions reference GA4 audiences.
- Finish → Done.
- In the list, click the account → Keys tab → Add Key → Create new key → JSON.
- Download the JSON file and place it in the project root as
serviceAccount.json
.
Note: Do not commit serviceAccount.json
to version control. It is already in .gitignore
.
3. Run the server
bun run index.ts
The server will start on port 3000 by default.
Usage
Add this MCP server to a client (e.g., Cursor, Claude Desktop, or your own MCP client)
In Cursor:
- Open Cursor Settings → Features → Add new MCP server.
- For the command, use:
npx -y supergateway --sse http://localhost:3000/mcp
"fire-config-mcp": {
"command": "npx",
"args": [
"-y",
"supergateway",
"--sse",
"http://localhost:3000/mcp"
]
}
(Or use the path/command as configured in your environment.)
- Save and connect.
In your own MCP client (TypeScript example):
You can connect to this server using the @modelcontextprotocol/sdk client:
import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { SSEClientTransport } from "@modelcontextprotocol/sdk/client/sse.js";
const client = new Client({ name: "my-client", version: "1.0.0" });
const transport = new SSEClientTransport("http://localhost:3000/mcp");
await client.connect(transport);
// Now you can list tools, call tools, etc.
const tools = await client.listTools();
For more details, see the MCP TypeScript SDK documentation.
This project was created using bun init
in bun v1.2.7. Bun is a fast all-in-one JavaScript runtime.