Firebase MCP for AI. Run backend ops without leaving your chat window.

Q: Can I update a document in Firestore without losing other fields?

Yes, you use firestorepatchdoc. This tool specifically modifies only the fields you name, leaving all other data intact. It prevents accidental overwrites.

Q: What is the difference between rtdbput and rtdbpatch?

rtdbput overwrites everything at a given path; it's like replacing the entire file. rtdbpatch, however, only updates specific children within that data structure.

Q: How do I send a message to all users?

You use fcmsendmessage. You specify the target topic or device list in your prompt, and the agent handles sending the notification payload through Firebase Cloud Messaging.

Q: After running authsignup, how do I retrieve the new user's profile data using authlookup?

The authlookup tool immediately fetches the profile details associated with a specific User ID (UID). You provide the UID, and the agent returns all stored profile information. This lets you verify the account creation status or pull existing metadata without needing to re-authenticate.

Q: If I need to check for multiple records in a collection, should I use firestorelistdocs instead of repeatedly calling firestoregetdoc?

Yes. Use firestorelistdocs when you want an overview of all documents within a whole collection; it returns IDs and basic data structure information for multiple records at once. You must use firestoregetdoc only if you already know the exact document ID you need to read.

Q: When deleting sensitive data, is there a functional difference between using rtdbdelete and firestoredeletedoc?

They target different database structures. rtdbdelete handles entire JSON paths in the Realtime Database, while firestoredeletedoc removes a single document from Cloud Firestore. Choose the tool based on whether your data lives in Firestore or RTDB.

Claude

ChatGPT

Cursor

Gemini

Windsurf

VS Code

JetBrains

Vercel

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

How this MCP server connects to your AI agent

Firebase REST & Admin APIs allows you to manage your backend data and user accounts directly from your AI agent.

Use this MCP to query Realtime Database and Firestore, sign up users, look up profiles, and send notifications using natural language commands.

What AI agents can do with Firebase (REST & Admin APIs) Automation

Auth lookup

Gets the profile details for a known Firebase Auth user.

Auth sign in

Signs in a user using their email and password credentials.

Auth sign up

Creates an account for a new user or signs them in anonymously.

+ 11 more capabilities included

Check user accounts

Look up existing user profiles or sign in new users using credentials.

Manage structured data (Firestore)

List, retrieve, and update entire collections of documents in Cloud Firestore.

Manipulate JSON data (RTDB)

Perform full read/write operations on complex, nested JSON structures within the Realtime Database.

Send notifications

Push targeted messages to specific devices or user groups using Firebase Cloud Messaging.

Ask an AI about this

Included with Plan

Waiting for input…

AI Agent

What AI agents can do with Firebase (REST & Admin APIs) MCP: 14 Tools

These tools give you granular control over every part of your Firebase backend, including user management, two different database types, and messaging services.

Make your AI actually useful.

Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.

Start using Firebase (REST & Admin APIs) on Vinkius

Auth Lookup

Gets the profile details for a known Firebase Auth user.

Auth Sign In

Signs in a user using their email and password credentials.

Auth Sign Up

Creates an account for a new user or signs them in anonymously.

Fcm Send Message

Sends a push message to specific devices or topics via Firebase Cloud Messaging.

Firestore Create Doc

Creates a brand new document in the specified Cloud Firestore collection.

Firestore Delete Doc

Deletes an entire document from Cloud Firestore.

Firestore Get Doc

Retrieves the data for a single, specific document in Cloud Firestore.

Firestore List Docs

Lists all documents currently present within a defined Firestore collection.

Firestore Patch Doc

Updates or inserts changes into an existing document in Cloud Firestore.

Rtdb Delete

Deletes a section of data from the Firebase Realtime Database.

Rtdb Get

Reads and retrieves all data stored at a specific path in the Realtime Database.

Rtdb Patch

Updates only specific pieces of data inside the Realtime Database structure.

Rtdb Push

Adds new data entries to a list or array within the Realtime Database.

Rtdb Put

Writes and overwrites all existing data at a specific path in the Realtime Database.

Security and governance baked right in.

Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.

Claude AI

Open Claude Settings

Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.

Add Custom Connector

Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:

https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp

Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. For OAuth-protected servers, expand Advanced settings to add credentials.

Start a conversation

Open a new chat. The Firebase integration is available immediately — no restart needed.

Antigravity

Configure Agent Environment

Open your Antigravity agent's workspace configuration or mcp-servers.json file.

Bind the Endpoint

Add the Vinkius endpoint URL to your agent's MCP connections list:

"mcp_servers": {
  "firebase-rest-admin-apis": {
    "serverUrl": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
  }
}

Provide your secure token in place of [YOUR_TOKEN_HERE] to ensure your agent requests are authenticated.

Execute

Start your Antigravity session. The agent will autonomously discover and utilize the Firebase tools with full Vinkius guardrails applied.

VS Code Copilot

⚡

One-Click Install (Recommended)

In your Vinkius Dashboard, simply click the Add to VS Code button for this server. We'll automatically configure your local workspace.

Or configure manually

Open MCP Settings

Open VS Code, press Ctrl/Cmd + Shift + P, and search for GitHub Copilot: MCP Servers.

Add Server Config

Add the Vinkius endpoint configuration to your mcp-servers.json file:

"firebase-rest-admin-apis": {
  "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
}

Ensure you replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.

LangChain

Install Dependencies

Install the LangChain MCP adapters for your environment:

pip install langchain-mcp-adapters

Connect the Server

Use the SSEClient in LangChain to connect to the Vinkius managed endpoint:

from langchain_mcp_adapters.client import SSEClient

# Connect to Vinkius
client = SSEClient(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")
tools = client.get_tools()

CrewAI

Define the Tool

Load the Vinkius MCP tools into your CrewAI agents:

from crewai import Agent
from mcp_crewai import MCPTool

# Connect securely to Vinkius
vinkius_tools = MCPTool(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")

# Assign to Agent
researcher = Agent(
    role='Data Researcher',
    tools=vinkius_tools.get_all()
)

Execute Task

Run your CrewAI process. The agent will autonomously route tasks to the Vinkius managed server.

Choose How to Get Started

Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.

Build Your Own

Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.

Import from OpenAPI, Swagger, or YAML specs
Create Agent Skills with progressive disclosure
Deploy to edge with MCPFusion framework
Built in DLP, auth, and compliance on every call
Real time usage dashboard and cost metering
Publish to catalog or keep private

Start building

Make Your AI Do More

Start with Firebase (REST & Admin APIs), then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.

Use this MCP plus 5,100+ others, all in one place
Add new capabilities to your AI anytime you want
Every connection is secured and compliant automatically
Track usage and costs across all your servers
Works with Claude, ChatGPT, Cursor, and more
New servers added to the catalog every week

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Firebase. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

VINKIUS INFRASTRUCTURE

Cloud Hosted

Managed infra

V8 Isolated

Sandboxed per request

Zero-Trust Proxy

No stored credentials

DLP Enforced

Policy on every call

GDPR Compliant

EU data residency

Token Compression

~60% cost reduction

Your data is protected. See how we built it.

Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more

The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.

This connection provides 14 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.

Checking your backend state used to mean context switching.

Today, if you need to verify user data or check application settings, the process is tedious. You have to log into the Firebase Console, navigate to Firestore, find the right collection, and then manually copy-paste IDs or run specific queries. If that's not enough, you might need to jump to a second console just for the Realtime Database view.

With this MCP, all of that goes away. You simply tell your agent what data you need—whether it's a user profile via `auth_lookup` or a collection list using `firestore_list_docs`. The agent handles the connections and spits out the clean answer right where you are working.

Manage users and notifications with Firebase (REST & Admin APIs)

Before, triggering an onboarding flow meant writing a script that first had to sign up the user (`auth_sign_up`), then write their initial data structure to RTDB, and finally schedule a welcome email. It was three separate steps across three different APIs.

Now, you just tell your agent: 'Onboard this new user.' The MCP coordinates signing them up, writing their initial state using `rtdb_put`, and sending the first notification with `fcm_send_message`. Everything happens in one flow.

Support 24/7 support@vinkius.com ↗

Security Vinkius Trust Center ↗

SLA Service Level Agreement ↗

Report Listing Send Report ↗

firebase

nosql

realtime-database

firestore

authentication

fcm

backend-as-a-service

What your AI can actually do with this

This connector gives your AI agent direct access to the core services powering a Firebase project. Instead of opening multiple developer consoles or writing boilerplate API calls, you tell your agent what needs fixing or checking. It handles reads and writes across both Cloud Firestore and Realtime Database structures. Need to verify if a user exists? Ask for it.

Want to push out an announcement? Send the message directly. This gives you complete control over authentication, data storage, and messaging—all without leaving your primary workflow. Vinkius hosts this MCP so you can connect it once from any compatible client and access all these backend tools instantly.

Built · Hosted · Managed by Vinkius Firebase APIs MCP - Manage Data & Users

Server ID 019e5d19-86a0-739c-8287-8730f03e930b

Vinkius Inspector

Compliance Grade F

Score 12.97/100

Report View Report ↗

Here's how it actually works

The bottom line is you get to manage complex backends using conversation, not console clicks.

Subscribe to this MCP and supply your specific Firebase Project ID and Web API Key.

Your AI client sends a natural language request, telling the agent what data needs reading or writing.

The agent executes the correct tool call (e.g., firestore_get_doc) against your live backend data and returns the result.

What Changes When You Connect

Debugging is faster. Instead of opening the console to check if a user exists, you simply ask your agent to run auth_lookup and get the profile details immediately.

Data updates are instant. Need to fix a setting? Use firestore_patch_doc or rtdb_patch to update specific fields right away, eliminating manual dashboard access.

Automation is simple. You can trigger complex sequences, like signing up a user with auth_sign_up, and then immediately sending them an onboarding message via fcm_send_message—all in one prompt.

View everything at once. Want to see what data lives in your app? Use firestore_list_docs or rtdb_get to inspect entire collections or paths instantly for status checks.

Complex state management is easier. You don't have to worry about cross-platform tool calls; the agent handles coordinating actions across Firestore, RTDB, and Auth services for you.

See it in action

01 01

A Product Manager needs live user stats.

The PM asks the AI client to list all documents in the 'user_metrics' collection. The agent runs firestore_list_docs, showing the latest activity counts immediately, letting them verify if a recent feature rollout worked.

02 02

A DevOps engineer needs to clean up old data.

The engineer realizes that user session logs are accumulating garbage. They ask the agent to delete all documents older than 90 days from the 'logs' collection using firestore_delete_doc.

03 03

A Backend Developer needs to test a new feature flow.

The developer simulates a user signing in with auth_sign_in, and then immediately uses rtdb_put to write the temporary session data to the Realtime Database for testing purposes.

04 04

A Support Agent needs to notify all affected users.

The agent confirms a service outage, then instructs the AI client to send an urgent alert using fcm_send_message to all targeted devices or topics across the user base.

The honest tradeoffs

Updating data without checking first.

Anti-pattern

The developer tries to overwrite a document entirely using a general API call, losing fields that were set by another service in the process.

The Fix

Don't use full write operations. Use firestore_patch_doc or rtdb_patch. These tools let you modify only specific children while leaving the rest of the data intact.

Assuming one tool handles all state.

Anti-pattern

Trying to run an operation that requires reading from Firestore, then writing to RTDB, but doing it in two separate steps with no coordination.

The Fix

The agent orchestrates this. You write the intent: 'First read X from Firestore, and if successful, update Y on RTDB.' The toolset allows these sequential actions.

Ignoring authentication requirements.

Anti-pattern

Attempting to firestore_create_doc for a user without first confirming their identity or credentials are valid.

The Fix

Always start by validating the user with auth_lookup. This ensures that any subsequent writes or operations use correctly identified and authorized data.

When It Fits, When It Doesn't

Use this MCP if your job involves managing live, stateful backend infrastructure. If you need to check a user's profile (auth_lookup), update database records across multiple services (Firestore and RTDB), or send automated notifications via FCM, this is the right tool. You are dealing with data that needs persistence and complex relationships.

Don't use it if you just need simple computation, like calculating sales tax on a number provided in the chat. For pure logic or basic text generation, your AI client has built-in capabilities. This MCP is for talking to your database; it isn't a replacement for code execution or math.

Questions you might have

How do I check if a user exists using auth_lookup? +

You provide the necessary unique ID to the agent, and it runs auth_lookup against Firebase Auth. It returns the profile data or an error if the account isn't found.

Can I update a document in Firestore without losing other fields? +

Yes, you use firestore_patch_doc. This tool specifically modifies only the fields you name, leaving all other data intact. It prevents accidental overwrites.

What is the difference between rtdb_put and rtdb_patch? +

rtdb_put overwrites everything at a given path; it's like replacing the entire file. rtdb_patch, however, only updates specific children within that data structure.

How do I send a message to all users? +

You use fcm_send_message. You specify the target topic or device list in your prompt, and the agent handles sending the notification payload through Firebase Cloud Messaging.

After running `auth_sign_up`, how do I retrieve the new user's profile data using `auth_lookup`? +

The auth_lookup tool immediately fetches the profile details associated with a specific User ID (UID). You provide the UID, and the agent returns all stored profile information. This lets you verify the account creation status or pull existing metadata without needing to re-authenticate.

If I need to check for multiple records in a collection, should I use `firestore_list_docs` instead of repeatedly calling `firestore_get_doc`? +

Yes. Use firestore_list_docs when you want an overview of all documents within a whole collection; it returns IDs and basic data structure information for multiple records at once. You must use firestore_get_doc only if you already know the exact document ID you need to read.

When deleting sensitive data, is there a functional difference between using `rtdb_delete` and `firestore_delete_doc`? +

They target different database structures. rtdb_delete handles entire JSON paths in the Realtime Database, while firestore_delete_doc removes a single document from Cloud Firestore. Choose the tool based on whether your data lives in Firestore or RTDB.

Before I start executing database commands, what core credentials must I provide to make this MCP functional? +

You must subscribe and input your Firebase Project ID and Web API Key. These two pieces of information grant the agent access rights to interact with your Realtime Database, Firestore, and Auth services.

Can I read data from a specific path in the Realtime Database? +

Yes. Use the rtdb_get tool and provide the database path (e.g., 'users/jack/profile'). You can also set the shallow parameter to true if you only want to see the keys at that level.

How do I create a new document in a Firestore collection? +

Use the firestore_create_doc tool. Specify the collectionPath and provide the document data as a JSON string. If you don't provide a documentId, Firestore will generate one automatically.

Can I check a user's profile information if I have their ID token? +

Absolutely. The auth_lookup tool allows you to retrieve full user profile data from Firebase Auth by providing a valid idToken obtained during sign-in.

How this MCP server connects to your AI agent

What AI agents can do with Firebase (REST & Admin APIs) Automation

Auth lookup

Auth sign in

Auth sign up

What AI agents can do with Firebase (REST & Admin APIs) MCP: 14 Tools

Make your AI actually useful.

Auth Lookup

Gets the profile details for a known Firebase Auth user.

Auth Sign In

Signs in a user using their email and password credentials.

Auth Sign Up

Creates an account for a new user or signs them in anonymously.

Fcm Send Message

Sends a push message to specific devices or topics via Firebase Cloud Messaging.

Firestore Create Doc

Creates a brand new document in the specified Cloud Firestore collection.

Firestore Delete Doc

Deletes an entire document from Cloud Firestore.

Firestore Get Doc

Retrieves the data for a single, specific document in Cloud Firestore.

Firestore List Docs

Lists all documents currently present within a defined Firestore collection.

Firestore Patch Doc

Updates or inserts changes into an existing document in Cloud Firestore.

Rtdb Delete

Deletes a section of data from the Firebase Realtime Database.

Rtdb Get

Reads and retrieves all data stored at a specific path in the Realtime Database.

Rtdb Patch

Updates only specific pieces of data inside the Realtime Database structure.

Rtdb Push

Adds new data entries to a list or array within the Realtime Database.

Rtdb Put

Writes and overwrites all existing data at a specific path in the Realtime Database.

Security and governance baked right in.

Claude AI

Open Claude Settings

Add Custom Connector

Start a conversation

Claude Code

Open your terminal

Add the MCP Server

Start coding

Cursor

One-Click Install (Recommended)

Open Cursor Settings

Add New Server

Use in Composer

Antigravity

Configure Agent Environment

Bind the Endpoint

Execute

VS Code Copilot

One-Click Install (Recommended)

Open MCP Settings

Add Server Config

Windsurf

One-Click Install (Recommended)

Open Windsurf Settings

Add Server Endpoint

LangChain

Install Dependencies

Connect the Server

CrewAI

Define the Tool

Execute Task

Choose How to Get Started

Build Your Own

Make Your AI Do More

Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more

Checking your backend state used to mean context switching.

Manage users and notifications with Firebase (REST & Admin APIs)

firebase

nosql

realtime-database

firestore

authentication

fcm

backend-as-a-service