Fireproof JSON Document Server MCP Server

DatabasesJavaScript

Store and query JSON documents with CRUD operations

Available Tools

createDocument

Creates a new JSON document in the database

document

getDocument

Retrieves a document by its ID

id

updateDocument

Updates an existing document by ID

iddocument

deleteDocument

Deletes a document by its ID

id

queryDocuments

Queries documents with optional sorting

sortBysortDirection

Fireproof JSON Document Server provides a simple yet powerful way to store, retrieve, update, and delete JSON documents directly from your AI assistant. Built on Fireproof database technology, it enables persistent document storage with the ability to query and sort documents by any field. This MCP server implements a complete set of CRUD operations (Create, Read, Update, Delete) for JSON documents, making it ideal for applications that need to maintain structured data across conversations. The integration with Model Context Protocol allows AI systems like Claude Desktop to interact with your data seamlessly.

Overview

Fireproof JSON Document Server is a Model Context Protocol (MCP) server that provides a JSON document database for AI assistants. It allows you to store, retrieve, update, and delete JSON documents, as well as query them with sorting capabilities.

Installation

To use the Fireproof JSON Document Server with Claude Desktop or other MCP-compatible AI assistants, follow these steps:

Clone the repository:

git clone https://github.com/fireproof-storage/mcp-database-server.git
cd mcp-database-server

Install dependencies and build the project:
```
npm install
npm run build
```
Configure your AI assistant to use the server by adding it to the appropriate configuration file:
- For Claude Desktop on macOS: Edit ~/Library/Application Support/Claude/claude_desktop_config.json
- For Claude Desktop on Windows: Edit %APPDATA%/Claude/claude_desktop_config.json
Add the following configuration, replacing /path/to/fireproof-mcp/build/index.js with the actual path to the built index.js file:
```
{
  "mcpServers": {
    "fireproof": {
      "command": "/path/to/fireproof-mcp/build/index.js"
    }
  }
}
```
Restart your AI assistant application to load the new MCP server.

Debugging

Since MCP servers communicate over stdio, debugging can be challenging. The repository includes integration with the MCP Inspector tool to help with debugging:

npm run inspector

This will provide a URL to access debugging tools in your browser, allowing you to monitor the communication between the AI assistant and the database server.

Usage Examples

Once installed, you can interact with the database through your AI assistant using natural language. Here are some example interactions:

"Create a new document with the title 'Meeting Notes' and content about our project discussion"
"Retrieve all documents sorted by creation date"
"Update the document with ID 'abc123' to add a 'priority' field set to 'high'"
"Delete the document with title 'Old Draft'"
"Find all documents with the tag 'important' sorted by title"

The server will handle these requests by translating them into the appropriate database operations and returning the results to your AI assistant.