Skip to content
MCP server

TerraVision

Draw AWS, Azure and GCP architecture diagrams from a prompt or Terraform code

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, macOS, Windows
Runs
Locally
Version
0.56.1
Listing updated
3 days ago

About TerraVision

TerraVision is an open-source MCP server for cloud and infrastructure. It runs locally on your machine and is started by your AI client.

TerraVision works in both directions. Describe a design in plain words and the assistant returns a diagram in PNG, SVG or editable draw.io form, which you can refine by asking for changes such as adding a cache or tracing a request. Ask for the Terraform and it is written from the diagram. Point it at existing Terraform, local or on GitHub, and it draws what terraform plan says will be deployed.

It runs on your own computer and needs no cloud account access, so a production architecture can be drawn without production credentials and before anything exists. The README says only metadata is sent to LLMs. Diagrams place resources in their VPC, subnet, zone or resource group using official icons, and can be turned into a single interactive HTML page showing per-resource settings.

The registry lists a stdio package on PyPI. Graphviz and Git must be installed, plus uv in most setups. The README marks AWS support as fullest, with Azure and GCP partial, and notes the project is alpha software provided as-is.

Install

Run in a terminal

claude mcp add terravision -- uvx terravision

Add to claude_desktop_config.json

{
  "mcpServers": {
    "terravision": {
      "command": "uvx",
      "args": [
        "terravision"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "terravision": {
      "command": "uvx",
      "args": [
        "terravision"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "terravision": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "terravision"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "terravision": {
      "command": "uvx",
      "args": [
        "terravision"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "terravision": {
      "command": "uvx",
      "args": [
        "terravision"
      ]
    }
  }
}

What it does

Prompt to diagram
Turn a plain-language design into an architecture diagram.
Terraform to diagram
Draw what Terraform will deploy without cloud access.
Diagram to Terraform
Write Terraform from a generated diagram.
Interactive HTML output
Click resources to see encryption, ports and IAM settings.
CI redraw
A GitHub Action redraws diagrams when Terraform changes.
Editable exports
PNG, SVG and draw.io formats.

Good for

  • Document existing Terraform for an architecture review
  • Sketch an AWS three-tier design and get its code
  • Keep a repository's diagram in sync in CI
  • Prepare an interactive view for a security audit

Pricing

Open source

Free and open source

Prices change. Check patrickchugh.github.io before you buy.

TerraVision FAQ

Does it need cloud credentials?

No, it works from code alone on your machine.

Which clouds are covered?

AWS most fully, with Azure and GCP partial per the README.

What must be installed?

Graphviz and Git, plus uv for most clients.

Is it free?

Yes, it is open source.

Is it production ready?

The README describes it as alpha software provided as-is.

Is TerraVision free?

Yes. TerraVision is open source, so you can use and self-host it for free.

How do I add TerraVision to Claude Code?

Run this command in your terminal: claude mcp add terravision -- uvx terravision

What are the best alternatives to TerraVision?

The closest alternatives to TerraVision are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to TerraVision

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools