firebase-mcp
Firebase MCP allows AI assistants to integrate with Firebase services such as Firestore, Storage, and Authentication. It supports MCP client applications like Claude Desktop and VS Code, providing robust file management and user verification services. Known issues have been identified, but do not affect functionality.
Overview
Firebase MCP enables AI assistants to work directly with Firebase services, including:
- Firestore: Document database operations
- Storage: File management with robust upload capabilities
- Authentication: User management and verification
The server works with MCP client applications like Claude Desktop, Augment Code, VS Code, and Cursor.
Known Issue: The firestore_list_collections tool may return a Zod validation error in client logs, but it does not affect functionality.
Quick Start
Prerequisites
- Firebase project with service account credentials
- Node.js environment
Installation
Configure your MCP settings and install the server via npx or local installation. Test the installation using an AI client.
Setup & Configuration
- Generate a Firebase service account key and set environment variables (required and optional).
- Integrate clients like Claude Desktop, VS Code/Augment, and Cursor with your configuration.
API Reference
- Firestore Tools: Add, list, get, update, delete documents, and query collections.
- Storage Tools: Manage files with listing, metadata retrieval, and uploads.
- Authentication Tools: Retrieve user information by ID or email.
Developer Guide
Installation and testing instructions are provided, including a project structure overview and troubleshooting common issues. Debugging tools and logging options are also covered.
HTTP Transport
Supports additional HTTP transport to run as a standalone service with session management.
Contributing
Guidelines for contribution include forking the repository, creating a feature branch, and ensuring test coverage.
License
MIT License - see LICENSE file for details.