Open-source prompt library, workflow hub, and site audit tool for vibe coders. Browse 52 battle-tested prompts, read 11 articles, follow a 9-step shipping playbook, and scan any landing page for issues in one click.
Live: https://vibeprompt.tech
vibeprompt is a free, open-source tool for developers shipping with AI.
- Browse 52 prompts organized by workflow stage, copy in one click
- Follow a 9-step vibe coding playbook from idea to shipped product
- Scan any site with PageLens, an instant SEO, conversion, security, and AI-readiness audit
- Explore a curated list of AI coding tools and resources
- No login required. No paywall.
| Route | What it does |
|---|---|
/ |
Homepage with workflow preview and FAQ |
/browse |
Browse all 52 prompts by category, with copy count tracking |
/prompts/[slug] |
Prompt detail with copy count and contributor badge |
/workflow |
9-step vibe coding playbook |
/scan |
PageLens site audit (SEO, conversion, trust, security, AEO) |
/articles |
Article index; /articles/[slug] for each post |
/awesome |
Curated list of AI/vibe coding tools |
/learn |
Learning resources |
/contact |
Contact |
/privacy |
Privacy policy |
/about |
About the project |
vibeprompt/
├─ src/
│ ├─ app/ # Next.js App Router pages
│ │ ├─ api/audit/ # PageLens audit API (40+ rules)
│ │ ├─ api/copy-counts/ # Prompt copy count API (Vercel KV)
│ │ └─ scan/ # PageLens scan UI
│ ├─ components/ # UI and feature components
│ │ ├─ prompts/ # Browse, cards, copy actions
│ │ ├─ layout/ # Navbar, footer
│ │ └─ ...
│ └─ lib/
│ ├─ actions/copies.ts # Copy count server actions (Vercel KV)
│ ├─ categories.ts # Slug to folder mappings (protected)
│ ├─ prompt-library.ts # Loads prompts from markdown
│ ├─ workflow-steps.ts # 9-step workflow metadata
│ └─ types.ts # Shared types (protected)
├─ content/ # Long-form content (e.g. articles)
├─ prompt-library/ # 52 public prompts (loaded at build time)
│ ├─ Agent Setup/
│ ├─ Architecture Stack/
│ ├─ Build Ship/
│ ├─ Launch Growth/
│ ├─ Ops Maintenance/
│ ├─ PRD Spec/
│ ├─ Prompting Craft/
│ ├─ Research Validate/
│ └─ Testing Quality/
├─ public/
│ ├─ llms.txt # AI crawler context file
│ └─ robots.txt # Explicit AI bot permissions
├─ AGENTS.md # Contributor and agent guide
└─ CLAUDE.md # Claude Code project instructions
| Layer | Technology |
|---|---|
| Framework | Next.js 16 (App Router) |
| UI | React 19, Tailwind CSS v4, shadcn/ui |
| Auth | None in the product (app is public, no sign-in) |
| KV storage | Vercel KV (copy counts and related use) |
| Database | Supabase client present, optional for future features |
| Animations | GSAP |
| Analytics | Vercel Analytics |
| Language | TypeScript (strict) |
npm install
npm run dev # localhost:3000
npm run lint
npm run typecheck
npm run buildNode 20+ (see package.json engines). Run commands from the repo root.
Environment variables needed in .env.local:
KV_REST_API_URL=
KV_REST_API_TOKEN=
# Optional:
NEXT_PUBLIC_SUPABASE_URL=
NEXT_PUBLIC_SUPABASE_ANON_KEY=52 markdown prompts in prompt-library/, organized by stage (folder names differ from on-site category labels, see src/lib/categories.ts):
| Folder | Site label | What it covers |
|---|---|---|
Agent Setup/ |
Context | AGENTS.md, CLAUDE.md, memory banks |
Research Validate/ |
Research | Validation, kill criteria, competitive analysis |
PRD Spec/ |
PRD | Requirements, acceptance criteria, scope |
Architecture Stack/ |
Stack | Stack decisions, system design |
Build Ship/ |
Build | Feature implementation, atomic tasks |
Prompting Craft/ |
Prompting | Prompt chains, output control |
Testing Quality/ |
Quality | Code review, security, testing |
Launch Growth/ |
Ship | Distribution, positioning, launch |
Ops Maintenance/ |
Iterate | Post-launch ops and reliability |
Adding a prompt
- Open
src/lib/categories.tsand use thedirNamefor the right folder underprompt-library/ - Add a kebab-case
.mdfile that matches the format inCONTRIBUTING.md(one fenced code block under## Instructions, placeholders listed under## Input, and so on) - Open a PR, for example:
prompt(category): add your-prompt-name
/scan audits any public URL across 40+ rules in 7 categories:
- SEO: title, meta description, canonical, headings
- Conversion: CTA presence, social proof, pricing, FAQ, email capture
- Trust: OG tags, structured data, privacy link, accessibility
- Structure: semantic HTML, viewport, nav/main/footer landmarks
- Performance: render-blocking scripts, lazy loading, resource hints
- Security: CSP, HSTS, X-Frame-Options, X-Content-Type-Options
- AI/AEO: llms.txt, FAQ schema, robots.txt AI crawler permissions
Audit API: GET /api/audit?url=https://example.com
Done
- 52 prompts across 9 categories, 11 articles, PageLens, workflow, copy counts, public access with no login
Next (see also ROADMAP.md and Issues)
- Community prompts and content (open issues: labels
content,good first issue) - Hardening copy counts and any infra follow-ups
- Ongoing quality: search, redirects (for example
wwwto apex), docs
For people: start with CONTRIBUTING.md. For agents and full stack detail: AGENTS.md.
- Quick contribution: copy fix, UI, mobile, add a prompt, add a tool, article tweak
- Roadmap:
ROADMAP.md - Issues: https://github.com/dotsystemsdevs/vibeprompt/issues
- PRs: https://github.com/dotsystemsdevs/vibeprompt/pulls
MIT, see LICENSE