Provides tools for querying live Firebase projects, including Firestore database management (listing collections, subcollections, and documents) and accessing Cloud Function logs with filtering capabilities.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Firebase Live MCP Server show me the last 10 error logs for the 'sendEmail' function"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for querying live Firebase projects - Firestore and Cloud Function logs.
npm install
npm run buildRelated MCP server: FireMCP
Uses Application Default Credentials. Run:
gcloud auth application-default loginThe server auto-detects the Firebase project from .firebaserc in the current working directory.
Override with env var:
FIREBASE_PROJECT_ID=my-projectAdd to claude_desktop_config.json:
{
"mcpServers": {
"firebase-live": {
"command": "node",
"args": ["/Users/nigelthorne/code/firebase_live_mcp_server/dist/index.js"],
"cwd": "/path/to/your/firebase/project",
"env": {
"FIREBASE_PROJECT_ID": "optional-override"
}
}
}
}The cwd should point to a directory with .firebaserc for auto-detection.
Tool | Description |
| List top-level Firestore collections |
| List subcollections of a document |
| List documents in a collection |
| Get a single document by path |
| Query with filters, ordering, limit |
| Get Cloud Function logs with grep filtering |
pattern - Regex to filter log messages
severity - Minimum level: DEBUG/INFO/NOTICE/WARNING/ERROR/CRITICAL
functionName - Filter by function name
limit - Max entries (default 50)
hoursAgo - Fetch logs from last N hours (default 1)
Your gcloud account needs:
roles/datastore.viewer - Firestore read access
roles/logging.viewer - Cloud Logging read access
This server cannot be deployed
Interact with your Google Cloud Firestore resources using natural language commands.
Enable secure connectivity between Sentry issues and debugging data, and LLM clients, using a Model Context Protocol (MCP) server.
Drive real devices from your AI Coding tool. Embed a client SDK (Unity, Godot, Flutter, iOS/macOS, Android, React Native, Web) in your app, then capture screenshots, traverse the UI tree, inject taps and key events, and run automated test tasks on the physical device over a secure relay.
Shared project context for AI agents and teams: docs, tasks, and messages that stay current.
