Firefly MCP Server

1[![Firefly](https://infralight-templates-public.s3.amazonaws.com/company-logos/firefly_logo_white.png)](https://firefly.ai)
2 
3# Firefly MCP Server
4 
5The Firefly MCP (Model Context Protocol) server is a TypeScript-based server that enables seamless integration with the Firefly platform. It allows you to discover, manage, and codify resources across your Cloud and SaaS accounts connected to Firefly.
6 
7## Features
8 
9- 🔍 Resource Discovery: Find any resource in your Cloud and SaaS accounts
10- 📝 Resource Codification: Convert discovered resources into Infrastructure as Code
11- 🔐 Secure Authentication: Uses FIREFLY_ACCESS_KEY and FIREFLY_SECRET_KEY for secure communication
12- 🚀 Easy Integration: Works seamlessly with Claude and Cursor
13 
14## Prerequisites
15 
16- Node.js (v20 or higher)
17- npm or yarn
18- Firefly account with generated access keys
19 
20## Installation
21 
22You can run the Firefly MCP server directly using NPX:
23 
24```bash
25npx @fireflyai/firefly-mcp
26```
27 
28### Environment Variables
29 
30You can provide your Firefly credentials in two ways:
31 
321. Using environment variables:
33 
34```bash
35FIREFLY_ACCESS_KEY=your_access_key FIREFLY_SECRET_KEY=your_secret_key npx @fireflyai/firefly-mcp
36```
37 
382. Using arguments:
39 
40```bash
41npx @fireflyai/firefly-mcp --access-key your_access_key --secret-key your_secret_key
42```
43 
44## Usage
45 
46### Stdio
47 
48Update the `mcp.json` file with the following:
49 
50```bash
51{
52  "mcpServers": {
53    "firefly": {
54      "command": "npx",
55      "args": ["-y", "@fireflyai/firefly-mcp"],
56      "env": {
57        "FIREFLY_ACCESS_KEY": "your_access_key",
58        "FIREFLY_SECRET_KEY": "your_secret_key"
59      }
60    }
61  }
62}
63```
64 
65Run the MCP server using one of the methods above with the following command:
66 
67```bash
68npx @fireflyai/firefly-mcp --sse --port 6001
69```
70 
71Update the `mcp.json` file with the following:
72 
73```bash
74{
75  "mcpServers": {
76    "firefly": {
77      "url": "http://localhost:6001/sse"
78    }
79  }
80}
81```
82 
83### Using with Cursor
84 
851. Start the MCP server using one of the methods above
862. Use the Cursor extension to connect to the MCP server - see [Cursor Model Context Protocol documentation](https://docs.cursor.com/context/model-context-protocol)
873. Use natural language to query your resources
88 
89#### Example:
90 
91##### Prompt
92 
93```
94Find all "ubuntu-prod" EC2 instance in 123456789012 AWS account and codify it into Terraform
95```
96 
97##### Response
98 
99```
100resource "aws_instance" "ubuntu-prod" {
101  ami           = "ami-0c55b159cbfafe1f0"
102  instance_type = "t3.micro"
103}
104```
105 
106## Demo
107 
108https://github.com/user-attachments/assets/0986dff5-d433-4d82-9564-876b8215b61e
109 
110## Contributing
111 
1121. Fork the repository
1132. Create your feature branch (`git checkout -b feature/amazing-feature`)
1143. Commit your changes (`git commit -m 'feat: Add amazing feature'`)
1154. Push to the branch (`git push origin feature/amazing-feature`)
1165. Open a Pull Request
117 
118## License
119 
120This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
121 
122## Support
123 
124For support, please visit [Firefly's documentation](https://docs.firefly.ai) or create an issue in this repository.
125