Firebase (REST & Admin APIs) MCP. Manage your database and user sessions via natural conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Firebase (REST & Admin APIs) MCP Server manages your entire backend infrastructure. It lets your AI agent query data in Realtime Database, manage records in Cloud Firestore, sign up users, or send FCM notifications—all through natural language commands.
Connect your agent to read, write, and control your entire Firebase project without leaving your editor.
What your AI agents can do
Auth lookup
Gets user data from Firebase Auth using a specific user ID.
Auth sign in
Signs in a user to Firebase using their email and password.
Auth sign up
Signs up a new user or allows an existing user to sign in anonymously.
The agent can sign up new users, authenticate existing ones, or look up profile details using the user's ID token.
The agent can list collections, retrieve specific documents, and perform full read, create, update, and delete operations on Firestore records.
The agent can read, write, patch, append, or delete specific data nodes within the Realtime Database JSON structure.
The agent sends Firebase Cloud Messaging (FCM) notifications to target devices or topics.
The agent retrieves specific user data from Firebase Authentication using a user ID.
The agent can create new documents in Firestore or write/overwrite specific data nodes in the Realtime Database.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Firebase (REST & Admin APIs) MCP Server: 14 Tools for Backend Ops
These 14 tools let your AI agent perform every necessary operation—from user sign-ins to reading and writing specific database nodes—across Firestore, RTDB, and Auth.
019e5d19auth lookup
Gets user data from Firebase Auth using a specific user ID.
019e5d19auth sign in
Signs in a user to Firebase using their email and password.
019e5d19auth sign up
Signs up a new user or allows an existing user to sign in anonymously.
019e5d19fcm send message
Sends a message notification via Firebase Cloud Messaging to specified devices or topics.
019e5d19firestore create doc
Creates a brand new document in a specified Cloud Firestore collection.
019e5d19firestore delete doc
Deletes a specific document from a Cloud Firestore collection.
019e5d19firestore get doc
Reads the content of a single, specified document in Cloud Firestore.
019e5d19firestore list docs
Lists all document IDs within a specified Cloud Firestore collection.
019e5d19firestore patch doc
Updates or inserts data into an existing document in Cloud Firestore.
019e5d19rtdb delete
Deletes data at a specified path within the Firebase Realtime Database.
019e5d19rtdb get
Reads data stored at a specific path in the Firebase Realtime Database.
019e5d19rtdb patch
Updates specific data nodes within the Firebase Realtime Database.
019e5d19rtdb push
Adds new data to a specified path in the Firebase Realtime Database, appending to the existing data.
019e5d19rtdb put
Writes data to a specific path in the Firebase Realtime Database, completely overwriting any existing data.
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
Make Your AI Do More
Start with Firebase (REST & Admin APIs), then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ 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
What you can do with this MCP connector
Yo, this Firebase APIs MCP Server hooks up your whole backend—Firestore, RTDB, and Auth—straight to your AI agent. You'll use it to manage everything using natural language commands. You can make your agent read, write, and control your entire Firebase project without ever leaving your editor.
User Accounts
Your agent can sign up new users using auth_sign_up, sign in existing ones using auth_sign_in, or check a user's profile details with auth_lookup if you give it the user ID token.
Cloud Firestore Data
You can use firestore_list_docs to list all document IDs in a collection. To get a specific document, you use firestore_get_doc. If you need to change something, you've got firestore_patch_doc to update or insert data into an existing record. To make a whole new record, just call firestore_create_doc. You can also wipe out data using firestore_delete_doc for a specific document.
How Firebase (REST & Admin APIs) MCP Works
- 1 Subscribe to the server and provide your Firebase Project ID and Web API Key.
- 2 Your AI client sends a request to the server, referencing the specific Firebase tool (e.g.,
firestore_get_doc). - 3 The server executes the tool call against your live Firebase project and returns the resulting data or status.
The bottom line is, your AI agent treats your entire Firebase backend like a set of exposed, natural language tools.
Who Is Firebase (REST & Admin APIs) MCP For?
Backend developers who hate context switching. DevOps engineers who spend too much time clicking through dashboards to check logs. Product managers who need live data status checks without needing a custom dashboard built. This is for anyone who needs to interact with live backend state directly from their coding environment.
Patches records in Firestore or tests Auth flows without leaving their IDE. They can run rtdb_get to inspect database state quickly.
Automates routine data maintenance tasks or verifies user credentials during an incident. They use the server to check user roles via auth_lookup.
Queries live user data for status checks or verification. They ask the agent to list documents in a collection using firestore_list_docs.
What Changes When You Connect
- Read live data instantly: Instead of navigating to the Firebase console, you ask your agent to
firestore_get_docorrtdb_getthe data. You get the JSON payload right in your chat window. - Automate user flow testing: Need to verify a user account? Use
auth_sign_upto create a test user, thenauth_lookupto check their profile data. It's all in one conversation. - Simplify data maintenance: Instead of writing complex console scripts, you tell the agent to
rtdb_patcha setting orfirestore_patch_doca record. It handles the API calls. - Streamline messaging: You can send out critical alerts using
fcm_send_messagewithout opening the Firebase console. Just tell your agent who needs the message. - Control the whole stack: The server gives you granular control. You can list collections with
firestore_list_docsthen create a document withfirestore_create_doc—all managed by the agent. - Write data safely: Whether you need to overwrite data (
rtdb_put) or just append logs (rtdb_push), the agent directs the exact operation you need, minimizing errors.
Real-World Use Cases
Verifying a Bug Fix
A dev finds a bug in the user profile display. Instead of manually logging into the console, they prompt their agent: 'Check the data at path user_profiles/test_user in the Realtime Database.' The agent runs rtdb_get and spits out the current JSON, letting them confirm the fix immediately.
Onboarding a New Feature
A PM wants to test a new feature requiring a new user. They instruct the agent to run auth_sign_up with test credentials. Next, they ask the agent to firestore_create_doc in the feature_flags collection, setting the status to 'active'. The entire sequence happens in chat.
Bulk Data Cleanup
A DevOps engineer needs to remove old, orphaned records. They ask the agent to first run firestore_list_docs on the old_logs collection, identify the stale IDs, and then run firestore_delete_doc for each one. No manual copy/paste needed.
Simulating a Push Notification
A developer wants to test the push notification service. They prompt the agent: 'Send a test message to the 'premium' topic.' The agent executes fcm_send_message, confirming the topic was targeted, allowing immediate verification of the messaging flow.
The Tradeoffs
Mixing concerns
Trying to update a user's name and then list all other users in the console. This forces two separate administrative workflows, increasing the chance of forgetting a step or using the wrong permission.
→
Use the agent to combine these steps. First, use firestore_patch_doc to update the specific user's name. Then, ask the agent to run firestore_get_doc for that user's ID to confirm the change in one flow.
Ignoring data types
Manually writing a script that assumes a field exists but doesn't handle missing data gracefully, causing the script to fail when a record is missing a key.
→
Use firestore_get_doc to safely read the document, then let the agent perform the update with firestore_patch_doc. This validates the document's existence and structure first.
Manual Auth Management
Having to switch between the Firebase Auth console and the database console every time a new user signs up, risking data inconsistency between the two services.
→
Let the agent handle the full loop. Run auth_sign_up for the new user, then immediately use firestore_create_doc to record their profile data, ensuring the user record is created immediately after successful authentication.
When It Fits, When It Doesn't
Use this server if your primary need is to treat your entire Firebase backend—Auth, Firestore, and RTDB—as a single, queryable system through natural language. It's perfect for developers and ops teams who need to test data flows, audit records, or manage users without leaving their IDE. Don't use this if you only need to interact with a single, non-Firebase service (like Stripe or GitHub). If your process involves multiple steps that touch different Firebase services (e.g., 'Create user, then write their profile, then send a welcome message'), this is the right tool. If you only need to read static data from a single source, a basic database connector might suffice, but this tool gives you the power to write and manage state.
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
Works with 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 server provides 14 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Manually checking user data or running CRUD operations shouldn't take a dozen clicks.
Right now, if you need to patch a single user's record or read a setting, you're clicking through the Firebase console. You jump to the Firestore tab, find the collection, open the document, and then use the update panel. If you need to check the user's Auth status too, you have to switch tabs and use a different UI.
With this MCP server, you just ask your agent. You prompt it to 'Update the user's email to X' or 'List all documents in the products collection.' The agent runs the necessary tool (`firestore_patch_doc` or `firestore_list_docs`) and gives you the result instantly, right in your conversation.
Firestore (REST & Admin APIs) MCP Server: Manage data and user sessions
Before, updating a record meant navigating to the database UI, finding the specific path, and manually typing the patch or overwrite values. If you missed a field, you'd get an API error and spend time debugging the console instead of the code.
Now, you talk to your agent. You tell it what to do—'Create a new document for the shipment.' The agent handles the `firestore_create_doc` call, ensuring the data structure is correct and the operation executes without you touching a single dashboard.
Common Questions About Firebase (REST & Admin APIs) MCP
How do I check a user's profile data using the auth_lookup tool? +
You pass the user's unique ID to the auth_lookup tool. The server returns the stored profile details, letting you confirm the user's status without needing the login password.
Can I use firestore_list_docs to see all documents in a collection? +
Yes, firestore_list_docs queries a collection and returns a list of all document IDs. You can then ask the agent to get the details for specific IDs from that list.
Is rtdb_get the same as firestore_get_doc? +
No. rtdb_get reads data from the Realtime Database, which is a separate service from Firestore. firestore_get_doc reads from the more modern Cloud Firestore. You use the tool that matches the data source you need.
How do I send a notification using fcm_send_message? +
You provide the target topic or device token to the fcm_send_message tool. The agent sends the message through Firebase Cloud Messaging, triggering the notification flow.
What's the difference between using `rtdb_put` and `rtdb_patch` to update data? +
rtdb_put overwrites the entire data structure at a path. rtdb_patch updates only the specified fields while leaving other data untouched. Use rtdb_patch when you only want to change a few values.
How do I manage user sign-ups or sign-ins using `auth_sign_up` and `auth_sign_in`? +
To sign up, call auth_sign_up with the user's email and password. You can then use auth_sign_in to authenticate them later. This flow handles both new user creation and subsequent logins.
Can I use `firestore_create_doc` or `firestore_patch_doc` if I don't specify a document ID? +
If you omit the document ID, both firestore_create_doc and firestore_patch_doc will automatically generate a unique ID for the new document. This is useful when you don't know the target ID beforehand.
What should I do if the `rtdb_get` or `firestore_get_doc` call fails due to permissions? +
If you get a permission error, check the service account credentials configured in your Vinkius Marketplace subscription. The service agent needs appropriate write and read permissions set in the Firebase console.
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.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Fourier Transform Engine
Calculate Fast Fourier Transforms (FFT) on time-series data local. Extract dominant frequencies from signals, audio, or financial cycles with mathematical perfection.
Kuaishou Mini-Game
Kuaishou mini-game developer API — manage cloud storage, leaderboards, analytics, and content moderation for casual games.
UUID Generator API
Generate unique identifiers — audit UUIDs and versions via AI.
You might also like
Fly.io Extended
Manage Fly.io apps, machines, and infrastructure—provision resources, control machine lifecycles, and manage volumes directly from any AI agent.
Flickr Photo Discovery
Universal photography intelligence — search millions of public photos via AI.
BoxLock
Secure package deliveries with smart lockbox technology that verifies carriers and prevents porch theft at the door.