Back to catalog
ADR Analysis MCP Server
AI server for analyzing Architectural Decision Records (ADR). Provides architectural insights, technology stack detection, security checks, and improved TDD workflow for development projects.
AI server for analyzing Architectural Decision Records (ADR). Provides architectural insights, technology stack detection, security checks, and improved TDD workflow for development projects.
Installation
NPM Global
npm install -g mcp-adr-analysis-server
NPX
npx mcp-adr-analysis-server
RHEL 9/10 Installer
curl -sSL https://raw.githubusercontent.com/tosin2013/mcp-adr-analysis-server/main/scripts/install-rhel.sh | bash
From Source
git clone https://github.com/tosin2013/mcp-adr-analysis-server.git
cd mcp-adr-analysis-server
npm install && npm run build && npm test
Configuration
Claude Desktop
{
"mcpServers": {
"adr-analysis": {
"command": "mcp-adr-analysis-server",
"env": {
"PROJECT_PATH": "/path/to/your/project",
"OPENROUTER_API_KEY": "your_key_here",
"EXECUTION_MODE": "full"
}
}
}
}
Features
- AI architecture insights with OpenRouter.ai integration
- Technology detection and architectural pattern identification
- ADR generation, suggestions, and support
- Smart code linking with AI-powered search
- Security and compliance with automatic content masking
- Test-Driven Development integration with two-phase validation
- Deployment readiness validation with zero-tolerance testing
- Tree-sitter AST analysis
- Web research via Firecrawl integration
- 95% confidence scoring for analysis results
Environment Variables
Required
OPENROUTER_API_KEY— API key for OpenRouter.ai integrationPROJECT_PATH— path to the project directory for analysisEXECUTION_MODE— set to 'full' to enable full analysis capabilities
Optional
FIRECRAWL_ENABLED— enable web research via FirecrawlFIRECRAWL_API_KEY— API key for the Firecrawl cloud serviceFIRECRAWL_BASE_URL— base URL for self-hosted Firecrawl instance
Usage Examples
Analyze this React project's architecture and suggest ADRs for any implicit decisions
Generate ADRs from the PRD.md file and create a todo.md with implementation tasks
Check this codebase for security issues and provide masking recommendations
Notes
Returns actual analysis results, not prompts to send elsewhere. Supports multiple AI assistants: Claude, Cline, Cursor, and Windsurf. Requires Node.js 20+ and TypeScript 5.9+. Includes >80% test coverage and comprehensive security features.
