Firefly MCP Server

Cloud PlatformsTypeScript

Discover, manage, and codify cloud resources across your infrastructure

Available Tools

findResources

Find resources in your Cloud and SaaS accounts based on search criteria

codifyResource

Convert discovered resources into Infrastructure as Code (Terraform, CloudFormation, etc.)

The Firefly MCP server enables seamless integration with the Firefly platform, allowing you to discover and manage resources across your Cloud and SaaS accounts. It provides powerful capabilities for finding any resource in your connected accounts and converting discovered resources into Infrastructure as Code, making it easier to manage your cloud infrastructure programmatically.

Overview

Firefly MCP server provides a bridge between AI assistants and your cloud infrastructure, allowing you to use natural language to discover, manage, and codify resources across your Cloud and SaaS accounts connected to Firefly.

Features

Resource Discovery: Find any resource in your Cloud and SaaS accounts
Resource Codification: Convert discovered resources into Infrastructure as Code
Secure Authentication: Uses Firefly access and secret keys for secure communication
Easy Integration: Works seamlessly with Claude and Cursor

Prerequisites

Before using the Firefly MCP server, you'll need:

Node.js (v14 or higher)
npm or yarn
A Firefly account with generated access keys

Installation

You can run the Firefly MCP server directly using NPX:

npx @fireflyai/firefly-mcp

Configuration

You can provide your Firefly credentials in two ways:

Using environment variables:

FIREFLY_ACCESS_KEY=your_access_key FIREFLY_SECRET_KEY=your_secret_key npx @fireflyai/firefly-mcp

Using command-line arguments:

npx @fireflyai/firefly-mcp --access-key your_access_key --secret-key your_secret_key

Running with Server-Sent Events (SSE)

To run the MCP server with SSE support:

npx @fireflyai/firefly-mcp --sse --port 6001

Using with AI Assistants

Once the server is running, you can use natural language to query your resources. For example:

Find all "ubuntu-prod" EC2 instances in 123456789012 AWS account and codify it into Terraform

The server will respond with the appropriate Terraform code:

resource "aws_instance" "ubuntu-prod" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t3.micro"
}

Support

For support, please visit Firefly's documentation or create an issue in the GitHub repository.