Architecture Review Agent

v1.0.0Coding Agentsstable

The Architecture Review Agent is an open-source AI agent sample that reviews software architectures and generates interactive diagrams automatically.

agent-architecturearchitecture-reviewazd-ai-templatesazure-openaidiagram-generation
Share:
74
Stars
0
Downloads
0
Weekly
0/5

What is Architecture Review Agent?

Architecture Review Agent is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to architecture review agent is an open-source ai agent sample that reviews software architectures and generates interactive diagrams automatically.

The Architecture Review Agent is an open-source AI agent sample that reviews software architectures and generates interactive diagrams automatically.

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • The Architecture Review Agent is an open-source AI agent sam

Use Cases

Automatically review software architectures and generate diagrams.
Create interactive visualizations of system designs.
Azure-Samples

Maintainer

LicenseMIT
Languagepython
Versionv1.0.0
UpdatedMay 20, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx agent-architecture-review-sample

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Architecture Review Agent

The Architecture Review Agent is an open-source Azure AI sample that automatically reviews software architecture documents and generates interactive diagrams. Given a YAML, Markdown, or plaintext description of a system, it parses components and connections, performs risk detection, maps relationships, and produces an Excalidraw diagram along with a structured review report. It demonstrates how to combine Azure OpenAI with the Model Context Protocol and draw.io or Excalidraw MCP tools for agentic diagram generation.

Prerequisites

  • Python 3.11 or later and pip installed
  • An Azure subscription with Azure OpenAI service enabled and a deployed model (e.g., gpt-4.1)
  • Azure OpenAI API key and endpoint URL
  • Microsoft Azure AI Foundry project endpoint (PROJECT_ENDPOINT)
  • Git installed for cloning the repository
1

Clone the repository

Clone the Azure Samples repository to your local machine.

git clone https://github.com/Azure-Samples/agent-architecture-review-sample
cd agent-architecture-review-sample
2

Run the setup script

Use the platform-specific setup script to create a virtual environment and install dependencies automatically.

# macOS/Linux
bash scripts/linux-mac/setup.sh

# Windows PowerShell
.\scripts\windows\setup.ps1
3

Configure environment variables

Copy the environment template and fill in your Azure OpenAI credentials and Foundry project endpoint.

cp .env.template .env
# Edit .env and set:
# PROJECT_ENDPOINT=https://your-project.services.ai.azure.com
# AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com/
# AZURE_OPENAI_API_KEY=your_api_key
# MODEL_DEPLOYMENT_NAME=gpt-4.1
4

Activate the virtual environment

Activate the Python virtual environment created by the setup script before running the agent.

# macOS/Linux
source .venv/bin/activate

# Windows
.\.venv\Scripts\Activate.ps1
5

Review an architecture file

Run the agent on a YAML or Markdown architecture description. The agent produces architecture.excalidraw, architecture.png, and review_bundle.json in the output/ directory.

python run_local.py scenarios/ecommerce.yaml
6

Review plaintext or force LLM inference

For unstructured text descriptions or to force LLM-based component extraction instead of rule-based parsing, use the --text or --infer flags.

# Plaintext architecture description
python run_local.py --text "Load Balancer -> Web Server -> PostgreSQL DB"

# Force LLM inference on any file
python run_local.py design_doc.txt --infer

Architecture Review Agent Examples

Client configuration

Environment variable configuration needed to run the Architecture Review Agent with Azure OpenAI.

{
  "PROJECT_ENDPOINT": "https://your-project.services.ai.azure.com",
  "AZURE_OPENAI_ENDPOINT": "https://your-resource.openai.azure.com/",
  "AZURE_OPENAI_API_KEY": "your_azure_openai_api_key",
  "MODEL_DEPLOYMENT_NAME": "gpt-4.1"
}

Prompts to try

Example commands and descriptions you can pass to the Architecture Review Agent.

- Run: python run_local.py scenarios/ecommerce.yaml
- Run: python run_local.py --text "API Gateway -> Auth Service -> User DB"
- Run: python run_local.py design_doc.md --infer --render
- In an MCP session: "Review the architecture described in scenarios/microservices.yaml and identify single points of failure."
- In an MCP session: "Generate an Excalidraw diagram from this architecture: CDN -> Load Balancer -> App Cluster -> Redis Cache -> PostgreSQL"

Troubleshooting Architecture Review Agent

Azure OpenAI returns 401 Unauthorized

Verify that AZURE_OPENAI_API_KEY matches the key shown in your Azure OpenAI resource under Keys and Endpoint, and that AZURE_OPENAI_ENDPOINT ends with a trailing slash.

SSL verification errors when behind a corporate proxy

Set ARCH_REVIEW_NO_SSL_VERIFY=1 in your .env file to disable SSL certificate verification. Do not use this in production environments.

output/ directory contains no diagram after running run_local.py

Check the terminal for parsing errors. If the architecture file uses a format the rule-based parser does not recognize, add the --infer flag to fall back to LLM-based extraction.

Frequently Asked Questions about Architecture Review Agent

What is Architecture Review Agent?

Architecture Review Agent is a Model Context Protocol (MCP) server that architecture review agent is an open-source ai agent sample that reviews software architectures and generates interactive diagrams automatically. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Architecture Review Agent?

Follow the installation instructions on the Architecture Review Agent GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Architecture Review Agent?

Architecture Review Agent works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Architecture Review Agent free to use?

Yes, Architecture Review Agent is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "agent-architecture-review-sample": { "command": "npx", "args": ["-y", "agent-architecture-review-sample"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Architecture Review Agent?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides