Luthersystems MCP server
{
"mcpServers": {
"luthersystems": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"docker.io/luthersystems/insideout-mcp:v0.36.3"
]
}
}
}{
"mcpServers": {
"luthersystems": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"docker.io/luthersystems/insideout-mcp:v0.36.3"
]
}
}
}{
"mcpServers": {
"luthersystems": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"docker.io/luthersystems/insideout-mcp:v0.36.3"
]
}
}
}{
"servers": {
"luthersystems": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"docker.io/luthersystems/insideout-mcp:v0.36.3"
]
}
}
}{
"mcpServers": {
"luthersystems": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"docker.io/luthersystems/insideout-mcp:v0.36.3"
]
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsonversion 0.36.3
What it does
InsideOut is an MCP server that designs, prices, and deploys cloud infrastructure on AWS and GCP based on natural language descriptions. It generates Terraform, provides cost estimates, and manages deployments directly from AI development tools, targeting solo developers, startups, and teams seeking faster infrastructure automation.
- Describe an application requirement in plain English and get a designed AWS/GCP architecture with Terraform generated and deployed within minutes.
- Get real-time cost estimates for proposed cloud infrastructure before deploying to understand budget impact.
- Inspect deployed environments, detect configuration drift, and manage resources over time with logging and monitoring built in from day one.
Tools this server exposes
| Tool | Does |
|---|---|
| design_architecture | Proposes a cloud architecture (services, networking, security, scaling) based on natural language application requirements. |
| estimate_cost | Calculates estimated monthly cloud costs for a proposed architecture before deployment. |
| generate_terraform | Produces production-ready Terraform configuration files from an architecture design. |
| deploy_infrastructure | Deploys the infrastructure directly to an AWS or GCP account and streams deployment logs. |
| inspect_environment | Queries deployed cloud infrastructure to report current state and configuration. |
| detect_drift | Identifies differences between expected and actual infrastructure configuration. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.