Vibe Techdesign

This is the third step in the vibe-coding workflow, right after you've written your PRD. It walks you through choosing your tech stack by asking questions based on your skill level: vibe-coder (AI does everything), experienced developer, or somewhere in between. Each path has different questions, from "free tools only or can you spend $50/month" to "monolithic or microservices architecture". After you answer, it generates a full technical design document with recommendations, alternatives, cost breakdowns, and a scaling path. The level-specific questioning is smart because it prevents overwhelming beginners with deployment strategies while still giving experienced developers the architectural depth they need. Saves to docs/TechDesign-[AppName]-MVP.md and tells you to run vibe-agents next.

skills.sh installs: 140
GitHub stars: 3.2K
last updated: Apr 19, 2026

View on GitHub

Install

$ npx -y skills add khazp/vibe-coding-prompt-template --skill vibe-techdesign --agent claude-code

Installs into .claude/skills of the current project.

Files

MVP technical design

Read the agreed requirements and reuse Handoff Context. Inspect an existing project's stack and relevant implementation before proposing replacements. Ask only for consequential choices that remain unresolved; a known answer does not need another confirmation echo.

Describe the architecture needed for the core journey: component/service boundaries, data ownership, integration contracts, deployment target, and relevant failure behavior. Prefer the smallest design that meets requirements. Add auth, storage, infrastructure, AI, or paid services only when justified. Verify changing vendor details from the installed code or official sources.

Record meaningful tradeoffs, compatibility constraints, migration/recovery requirements where applicable, and how the result will be checked. Distinguish local test operations from external sends, production writes, and deployments. Keep secrets out of generated documents. Do not assign broad tool permissions or require a team of agents to execute an ordinary implementation.

SKILL.md on GitHub