Project Architecture Planner
A cloud-agnostic architecture planner that runs a discovery interview, then delivers a tech stack evaluation, a phased scalability roadmap, a cost model and Mermaid/HTML/draw.io diagrams.
Overview
This agent plans software architecture end to end, from a greenfield idea to an existing codebase that needs direction. It's explicitly cloud agnostic and framework agnostic, and it does not generate application code: the output is plans, diagrams and cost models.
How it works
- It asks about business context, expected scale, team size and budget before recommending anything, adjusting how deep it goes based on whether the project is a simple app, growth-stage or enterprise-scale.
- It recommends at least two architecture styles (monolith, modular monolith, microservices, serverless, event-driven) with their trade-offs, rather than a single default.
- Tech stack candidates are scored on team fit, ecosystem maturity, scalability, cost and hiring market, with a primary and alternative choice per layer (frontend, backend, database...).
- The roadmap phases the plan across MVP, growth (1K to 100K users) and scale (100K+), specifying what changes and why at each stage.
- It produces a cloud-agnostic monthly cost model per phase and a multi-cloud comparison when relevant.
- It generates Mermaid diagrams embedded in markdown, an interactive HTML preview page, and a draw.io export for every plan.
Examples
"Plan the architecture for a B2B SaaS expecting 500 users at launch
and 50k within two years."
Installation
You need VS Code with the GitHub Copilot extension.
- Create the agents folder and download the file into your repo:
mkdir -p .github/agents curl -o .github/agents/project-architecture-planner.agent.md \ https://raw.githubusercontent.com/github/awesome-copilot/main/agents/project-architecture-planner.agent.md - Reload VS Code, or open a new Copilot Chat; custom agents under
.github/agents/are picked up automatically. - Pick this agent from the mode picker in Copilot Chat and describe a project to confirm it interviews you before proposing a stack.
Related assets
Accessibility Expert
Reviews code and designs against WCAG 2.1/2.2, covering semantics, keyboard and focus behavior, forms, media and dynamic SPA updates, and ships framework-specific examples plus a CI setup for automated checks.
Address Comments Agent
Works through pull request review comments one at a time: applies the minimal fix for each, pushes back on comments that don't make sense, adds test coverage, and commits with a descriptive message before moving to the next comment.
ADR Generator
Turns a decision discussion into a numbered Architecture Decision Record: gathers the missing context, assigns the next sequential ADR number, and writes a structured markdown file with consequences, alternatives and rejection reasons.