vibe-provision
Enables AI agents to provision and manage external SaaS services (Clerk, Stripe, Resend, etc.) by authenticating and creating resources via YAML configuration.
README
vibe-provision
Provision external SaaS services from YAML. One command to set up Clerk, Stripe, Resend and inject .env.
"AI can write code, but it can't click dashboards." — vibe-provision solves that.
Quick Start
# 1. Generate a config template
npx vibe-provision init
# 2. Authenticate with providers (one-time)
npx vibe-provision auth
# 3. Provision resources and generate .env
npx vibe-provision up
That's it. Your .env is ready — run your dev server.
vp is a short alias: npx vp up works too.
vibe.yaml
project: my-saas-app
output:
- .env
- vercel # auto-inject env vars to Vercel
- terraform # generate terraform.tfvars.json
services:
auth:
provider: clerk
config:
app_name: "My SaaS App"
redirect_urls:
- http://localhost:3000/callback
payments:
provider: stripe
config:
products:
- name: "Pro Plan"
prices:
- amount: 1900
currency: usd
interval: month
webhooks:
events:
- checkout.session.completed
- customer.subscription.updated
email:
provider: resend
config:
domain: my-app.com
database:
provider: neon
config:
region: aws-ap-northeast-1
cache:
provider: upstash
config:
region: ap-northeast-1
AI (Cursor, Claude Code, etc.) can generate this file alongside your app code.
Supported Providers
| Provider | Category | What it creates | Auth method |
|---|---|---|---|
| Clerk | Auth | Redirect URL config + env vars | API key paste |
| Stripe | Payments | Products, Prices, Webhook Endpoints | API key paste |
| Resend | Domain registration | API key paste | |
| Supabase | DB + Auth | Project + API keys | Access token |
| Neon | Postgres | Project + database | API key |
| Upstash | Redis | Database | Email + API key |
Output Targets
Control where env vars are written via the output section:
| Target | Description |
|---|---|
.env |
Local .env file (default) |
vercel |
Vercel environment variables via CLI |
terraform |
.vibe-provision/terraform.tfvars.json with merge semantics |
Environment-Specific Config
Use --env to manage multiple environments:
npx vp up --env dev # merges vibe.yaml + vibe.dev.yaml → .env.dev
npx vp up --env staging # merges vibe.yaml + vibe.staging.yaml → .env.staging
npx vp up --env prod # merges vibe.yaml + vibe.prod.yaml → .env.prod
npx vp up # uses vibe.yaml only → .env
Base config (vibe.yaml) holds shared settings. Override files (vibe.{env}.yaml) deep-merge on top:
# vibe.dev.yaml — only override what differs
services:
payments:
provider: stripe
config:
webhooks:
url: https://dev.example.com/api/webhooks/stripe
MCP Server (AI Agent Integration)
vibe-provision includes an MCP server so AI agents (Claude Code, Cursor) can provision services directly.
Setup
Add to your .mcp.json (global or per-project):
{
"mcpServers": {
"vibe-provision": {
"command": "npx",
"args": ["vibe-provision", "mcp"]
}
}
}
Available Tools
| Tool | Description |
|---|---|
vibe_provision_status |
Check auth and provisioning state for all providers |
vibe_provision_up |
Provision resources and generate .env (requires prior auth) |
vibe_provision_add |
Add a new service to vibe.yaml |
Example Flow
User: "Add Stripe payments to my app"
→ AI generates vibe.yaml with stripe config
→ AI calls vibe_provision_status → "stripe: NOT authenticated"
→ AI: "Run npx vp auth in your terminal"
→ User authenticates (one-time)
→ AI calls vibe_provision_up → Products, Prices, Webhooks created
→ .env updated, app ready to run
Idempotency
vibe-provision up is safe to run multiple times. It tracks created resources in .vibe-provision/state.json and skips anything that already exists.
How It Works
init— generates avibe.yamltemplateauth— walks you through authenticating each provider, stores credentials locally in~/.vibe-provision/auth/up— readsvibe.yaml, calls provider APIs to create resources, writes to configured output targets
Credentials never leave your machine.
Examples
- saas-starter-simple — Next.js + Clerk + Stripe + Resend, direct webhook handling
- saas-starter — Same stack + qhook for production webhook processing
Development
npm install
npm run lint # type check
npm test # run tests (47 tests)
npm run dev -- init # run CLI in dev mode
License
FSL-1.1-Apache-2.0 — Free to use for any purpose except competing hosted services. Converts to Apache 2.0 on 2028-03-26.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.