$ loading_
Guide AI coding with requirements, design, and task planning before implementation.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Universal Spec Architect" yet — see the docs or source repo.
I want to build a team task board web app. Follow a strict spec-driven workflow: produce requirements, system design, and task breakdown first, and do not write any code until I approve them.
A structured requirements doc, technical design, and actionable task list, with no code yet.
I want to refactor an existing payment module. First analyze the current state, then list functional requirements, non-functional requirements, risks, API changes, and a phased implementation plan before any coding.
A review-ready refactoring spec that aligns the team on approach and risks before coding.
Act as a spec architect for an AI coding agent. For any implementation task, always provide requirements, design decisions, acceptance criteria, and a task plan first; if information is missing, ask clarifying questions instead of generating code.
A standardized pre-implementation document with clarifying questions, acceptance criteria, and execution steps for later coding.
Turn coding requests into verifiable plan-act-verify loops with durable repo evidence.
Create, review, and manage specs for AI-driven coding and testing workflows.
Enable AI to read and write SpecDown specs for spec-driven development.
Refine product specifications through multi-model debate until consensus is reached.
Check whether project code matches SPEC documents using deterministic static analysis.
Add engineering discipline to AI coding assistants across design, implementation, and review.