Run a structured architecture interview before starting a new software project.
Generate Domain-Driven Design and CQRS documentation from a guided Q&A process.
Resume an interrupted architecture planning session exactly where you left off.
Feed an existing project's legacy documentation in to pre-fill architectural drafts.
| galego/sdd-interview | 000madz000/payload-test-api-route-handler | 0marildo/imago | |
|---|---|---|---|
| Stars | 3 | 3 | 3 |
| Language | — | TypeScript | Python |
| Setup difficulty | easy | easy | easy |
| Complexity | 2/5 | 2/5 | 2/5 |
| Audience | developer | developer | general |
Figures from each repo's GitHub metadata at analysis time.
No backend or infrastructure needed, but the full interview across all phases takes real time to complete.
SDD-Interview is a skill for AI coding assistants that acts like a senior software architect interviewing you before any code gets written. Instead of jumping straight into building a project, it walks a developer, technical lead, or founder through a structured series of questions, one at a time, and turns the answers into a full set of architectural documentation saved as Markdown files inside your project. The README describes the AI adopting the persona of a strict, analytical chief architect that asks one question at a time so it stays focused and does not lose track of context. The process is not a rigid script though: you can change your mind mid interview, and the AI is able to go back and edit documents it already generated earlier without breaking the overall flow. As the interview moves through different phases, the AI also cross references earlier answers, so a decision made about the database early on stays consistent with interface requirements decided later. The output covers a wide range of established software planning methods, including Domain-Driven Design, Command Query Responsibility Segregation, Clean Architecture, Event-Driven Development, Behavior-Driven Development, and Test-Driven Development, among others, organized into eighteen phases producing over thirty separate documents. The whole system tracks its own progress in a state file, so you can pause partway through and resume later without losing what was already decided. It is described as platform and model agnostic, meaning it is built only from Markdown files and prompt instructions rather than any custom software, so it can be used with different AI tools and different underlying language models. It requires no backend infrastructure and runs entirely inside your IDE or command line agent through text files, at no extra cost beyond the AI tool you are already using. It can also read an existing project's documentation to prefill some answers automatically, and the interview and its output documents can be conducted in different languages.
An AI-driven skill that interviews you like a chief software architect and produces a full set of Markdown architecture documents before you write any code.
MIT license: use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.