AI-powered architecture documentation framework for software projects. Generates ADRs, conducts reviews, and offers pragmatic mode. Integrates with Claude Code Plugin for easy installation. Benefits operations teams by automating architecture documentation and reviews.
git clone https://github.com/codenamev/ai-software-architect.gitAI Software Architect is a markdown-based framework that helps teams implement rigorous software architecture practices through structured documentation, reviews, and progress tracking. It provides templates and processes for creating Architecture Decision Records (ADRs), conducting multi-perspective architecture reviews, and translating feedback into implementation plans. The framework integrates seamlessly with Claude Code, Cursor, and other AI coding assistants via three installation methods: Claude Code Plugin (recommended for simplest setup), MCP Server (for programmatic automation), or traditional setup (for maximum flexibility). It supports a pragmatic mode for practical decision-making and offers advanced project analysis capabilities across all installation paths.
Install via Claude Code Plugin with two commands (`/plugin marketplace add` and `/plugin install`), or via npm for MCP Server integration, or use traditional setup by referencing the GitHub repository. The framework provides templates and processes accessible through your AI assistant for creating ADRs, conducting reviews, and monitoring implementation.
Generate and manage Architecture Decision Records (ADRs) automatically
Conduct multi-perspective architecture reviews with specialized reviewers
Track implementation progress of architectural changes across projects
Document architectural decisions in markdown for version control
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/codenamev/ai-software-architectCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Act as an AI software architect for our [PROJECT_NAME] project. Generate an Architecture Decision Record (ADR) for our decision to use [TECHNOLOGY] as our primary database. Include the context, decision, status, and consequences. Also, conduct a quick review of our current architecture focusing on [SPECIFIC_CONCERN] and provide recommendations.
Based on your request, I've generated an ADR for the decision to use MongoDB as the primary database for your e-commerce platform project. The context highlights the need for scalability and flexible data modeling to handle diverse product catalogs and customer data. The decision to use MongoDB is marked as 'Accepted' with the consequences being improved scalability and easier schema evolution, but with potential trade-offs in transactional integrity. For the architecture review, I've identified that your current microservices architecture could benefit from implementing a more robust service mesh to handle inter-service communication more efficiently. I recommend evaluating Istio or Linkerd for this purpose, as they provide advanced traffic management, security, and observability features.
AI assistant built for thoughtful, nuanced conversation
Get more done every day with Microsoft Teams – powered by AI
Automate security compliance and monitor real-time security posture seamlessly.
Automate your spreadsheet tasks with AI power
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan