AI Infrastructure Agent

Intelligent AWS infrastructure management through natural language. Build, modify, and manage your cloud resources with the power of AI.

Get Started Documentation

Get Started in Seconds

Run with Docker - just set your API keys and go

docker-run.sh

Shell

docker run -d \
  -p 8080:8080 \
  -e GEMINI_API_KEY="your-gemini-api-key-here" \
  -e AWS_ACCESS_KEY_ID=$AWS_ACCESS_KEY_ID \
  -e AWS_SECRET_ACCESS_KEY=$AWS_SECRET_ACCESS_KEY \
  -e AWS_DEFAULT_REGION=$AWS_DEFAULT_REGION \
  --name ai-agent \
  ghcr.io/versuscontrol/ai-infrastructure-agent

Access your browser

The web dashboard allows you to manage your infrastructure using natural language commands

Main Features

Everything you need to manage AWS infrastructure with AI

Natural Language Interface

Describe your desired infrastructure in plain English, and let the AI handle the rest.

Multi-AI Provider Support

Flexibly switch between OpenAI, Google Gemini, Anthropic Claude, and AWS Bedrock Nova.

Visual Web Dashboard

Manage and monitor your infrastructure through an intuitive web interface.

Terraform-like State

Maintains an accurate state of your infrastructure to prevent drift.

Conflict Detection

Safely execute plans with built-in conflict detection and dry-run mode.

Broad AWS Resource Support

Support for VPC, EC2, Security Groups, Auto Scaling Groups, and ALBs.

Supported AI Models

Choose the AI provider that best fits your needs and infrastructure requirements

OpenAI

Google Gemini

Anthropic Claude

AWS Bedrock Nova

OpenAI Setup Gemini Setup Bedrock Setup

The Sample

See how the AI agent transforms your natural language request into a detailed execution plan

Prompt

"Create an EC2 for hosting an Apache Server with a dedicated security group that allows inbound HTTP (port 80) and SSH (port 22) traffic."

Result

The agent creates a detailed execution plan to provision an EC2 instance and a corresponding security group with the specified rules, then awaits your approval before applying the changes.

For Example

Below is an interactive execution plan that shows how the AI agent breaks down your request into actionable steps, provides reasoning, and awaits your confirmation before making any changes to your infrastructure.

Request Processed Create an EC2 instance for hosting an Apache Server with a dedicated security group that allows inbound HTTP (port 80) and SSH (port 22) traffic.

Mode: Live Execution

Confidence: 95.0%

Reasoning

The user wants to create an EC2 instance with a dedicated security group. I need to discover a VPC, create a security group that allows HTTP and SSH traffic, and then create the EC2 instance using a suitable AMI. Since no managed resources are provided, I will discover the default VPC and then proceed with creating the security group and the EC2 instance.

Execution Plan