fireproof-storage/mcp-database-server
FreeFireproof 官方集成，不可变账本数据库，支持实时同步。
FreeFree tier
About fireproof-storage/mcp-database-server
This is a demo MCP (Model Context Protocol) server that integrates Fireproof, an immutable ledger database with real-time sync, to store and load JSON documents from LLM tool use. It implements basic CRUD operations (Create, Read, Update, Delete) and supports querying documents sorted by any field. Designed to plug code and data into AI systems such as Claude Desktop, it showcases how Fireproof can be used as a persistent storage layer for AI agents.
Key Features
CRUD operations (Create, Read, Update, Delete) for JSON documents
Query documents sorted by any field
Integration with Model Context Protocol (MCP) for AI tools like Claude Desktop
Leverages Fireproof's immutable ledger with real-time sync
Simple setup with npm install and build
Pros & Cons
Pros
- Leverages Fireproof's immutable ledger for data integrity and real-time sync
- Simple CRUD API familiar to developers
- Easy to integrate with Claude Desktop and other MCP-compatible AI systems
- Open source and free to use
- Good starting point for MCP server development
Cons
- This is a demo server and may not be production-ready
- Limited documentation beyond the basic README
- Requires Node.js and npm to run
- Debugging via stdio requires additional tools like MCP Inspector
Best For
Persistent storage for AI agent conversations and tool outputsDemo and prototyping for MCP-based applicationsStoring and retrieving JSON data from LLM interactionsBuilding AI applications with a real-time synchronized database
FAQ
What is the Model Context Protocol (MCP)?
MCP is a protocol used to plug code and data into AI systems such as Claude Desktop. This server implements an MCP server for Fireproof database.
How do I install and run this server?
Clone the repository, run 'npm install' and 'npm build', then configure your Claude Desktop config file to point to the built index.js.
What operations are supported?
The server supports CRUD operations (Create, Read, Update, Delete) on JSON documents and allows querying documents sorted by any field.
Is this server production-ready?
It is described as a simple demo/example, not yet a production-ready application.