Skip to content
Back to skills

Replit Multi Env Setup

ASecurity

'Configure Replit dev/staging/production environments with separate databases,

  • 2,691 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 2, 2026
ai-agentstypescriptsqlnodegitapidatabase

Works with

  • claude code
  • api

Security analysis

A100/100

Scanned September 2, 2026

npx -y skills add jeremylongshore/tons-of-skills-marketplace --skill replit-multi-env-setup --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Replit Multi Env Setup?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Replit Multi Env Setup
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jeremylongshore-replit-multi-env-setup-tons-of-skills-marketplace/badge)](https://www.skillsdirectory.com/skills/jeremylongshore-replit-multi-env-setup-tons-of-skills-marketplace)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: replit-multi-env-setup
description: 'Configure Replit dev/staging/production environments with separate databases,
  secrets, and deployment tiers.

  Use when setting up multi-environment deployments, managing per-environment secrets,

  or implementing environment-specific Replit configurations.

  Trigger with phrases like "replit environments", "replit staging",

  "replit dev prod", "replit environment setup", "replit separate databases".

  '
allowed-tools: Read, Write, Edit
version: 1.13.0
license: MIT
author: Jeremy Longshore <jeremy@intentsolutions.io>
tags:
- saas
- replit
- deployment
- environments
compatibility: Designed for Claude Code
---
# Replit Multi-Environment Setup

## Overview

Configure development, staging, and production environments on Replit. Leverages Replit's built-in dev/prod database separation, environment-specific secrets, and deployment types. Covers the Replit-native approach (single Repl, dual databases) and the multi-Repl approach (separate Repls per environment).

## Prerequisites

- Replit Core or Teams plan (deployment access)
- PostgreSQL provisioned in Database pane
- Understanding of Replit Secrets

## Environment Strategy

### Approach 1: Single Repl, Dual Databases (Recommended)

Replit natively provides separate development and production databases:

```markdown
Workspace "Run" button → Development database
Deployed app (.replit.app) → Production database

Both use the same DATABASE_URL env var — Replit routes automatically.
No code changes needed between environments.
```

### Approach 2: Multi-Repl (Staging + Production)

For teams that need a staging environment:

```markdown
Repl 1: my-app-staging → Autoscale deployment → staging.replit.app
Repl 2: my-app-prod   → Reserved VM deployment → app.example.com

Each Repl has its own:
- Secrets (different API keys per environment)
- PostgreSQL database (separate data)
- Deployment configuration
- GitHub branch (staging → staging, main → production)
```

## Instructions

### Step 1: Environment Detection

```typescript
// src/config/environment.ts
type Environment = 'development' | 'staging' | 'production';

export function detectEnvironment(): Environment {
  // Replit deployment context
  if (process.env.REPL_DEPLOYMENT) {
    // Check if this is the staging Repl
    if (process.env.REPL_SLUG?.includes('staging')) return 'staging';
    return 'production';
  }

  // Workspace "Run" context
  if (process.env.REPL_SLUG) return 'development';

  // Fallback to NODE_ENV
  const env = process.env.NODE_ENV || 'development';
  if (env === 'production') return 'production';
  if (env === 'staging') return 'staging';
  return 'development';
}

export const ENV = detectEnvironment();
export const IS_PROD = ENV === 'production';
```

### Step 2: Environment-Specific Configuration

```typescript
// src/config/index.ts
import { ENV, IS_PROD } from './environment';

const configs = {
  development: {
    logLevel: 'debug',
    corsOrigins: ['*'],
    rateLimit: { windowMs: 60000, max: 1000 },
    cache: { ttlMs: 5000 },
    features: { debugEndpoints: true },
  },
  staging: {
    logLevel: 'info',
    corsOrigins: ['https://staging.example.com'],
    rateLimit: { windowMs: 60000, max: 200 },
    cache: { ttlMs: 30000 },
    features: { debugEndpoints: true },
  },
  production: {
    logLevel: 'warn',
    corsOrigins: ['https://app.example.com'],
    rateLimit: { windowMs: 60000, max: 100 },
    cache: { ttlMs: 300000 },
    features: { debugEndpoints: false },
  },
};

export const config = {
  env: ENV,
  port: parseInt(process.env.PORT || '3000'),
  ...configs[ENV],
  db: {
    // DATABASE_URL auto-switches between dev and prod on Replit
    url: process.env.DATABASE_URL,
  },
};

// Validate production secrets
if (IS_PROD) {
  const required = ['DATABASE_URL', 'JWT_SECRET'];
  const missing = required.filter(k => !process.env[k]);
  if (missing.length) {
    console.error(`FATAL: Missing production secrets: ${missing.join(', ')}`);
    process.exit(1);
  }
}
```

### Step 3: Separate Secrets Per Environment

```markdown
For Single Repl (dev/prod separation):
- All secrets set once in Secrets tab
- DATABASE_URL auto-switches (Replit manages)
- Same JWT_SECRET for both (or use REPL_IDENTITY for dev)

For Multi-Repl (staging + prod):
- Each Repl has its own Secrets tab
- staging Repl: JWT_SECRET=staging-secret, API_KEY=test-key
- production Repl: JWT_SECRET=prod-secret, API_KEY=live-key

Account-level secrets (shared across all Repls):
- Settings > Secrets > Account secrets
- Useful for: monitoring tokens, shared infrastructure keys
```

### Step 4: GitHub Branch Strategy (Multi-Repl)

```markdown
Repository setup:
- main branch → connected to production Repl
- staging branch → connected to staging Repl
- feature branches → PR to staging first

Workflow:
1. Feature branch → PR to staging
2. Tests pass + review → merge to staging
3. Staging Repl auto-deploys → verify staging
4. Staging → PR to main
5. Merge → production Repl auto-deploys
```

```yaml
# .github/workflows/deploy.yml
name: Deploy Pipeline

on:
  push:
    branches: [staging, main]

jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - run: npm ci && npm test

  verify-staging:
    if: github.ref == 'refs/heads/staging'
    needs: test
    runs-on: ubuntu-latest
    steps:
      - name: Wait for Replit deploy
        run: sleep 60
      - name: Health check staging
        run: curl -sf ${{ secrets.STAGING_URL }}/health

  verify-production:
    if: github.ref == 'refs/heads/main'
    needs: test
    runs-on: ubuntu-latest
    steps:
      - name: Wait for Replit deploy
        run: sleep 60
      - name: Health check production
        run: curl -sf ${{ secrets.PRODUCTION_URL }}/health
```

### Step 5: Database Migration Between Environments

```typescript
// scripts/promote-data.ts — Copy staging data to production (carefully!)
import { Pool } from 'pg';

const staging = new Pool({ connectionString: process.env.STAGING_DATABASE_URL });
const production = new Pool({ connectionString: process.env.PRODUCTION_DATABASE_URL });

async function promoteConfig() {
  // Only promote configuration/reference data, never user data
  const { rows: configs } = await staging.query('SELECT * FROM feature_flags');

  for (const config of configs) {
    await production.query(
      `INSERT INTO feature_flags (key, value, updated_at)
       VALUES ($1, $2, NOW())
       ON CONFLICT (key) DO UPDATE SET value = $2, updated_at = NOW()`,
      [config.key, config.value]
    );
  }

  console.log(`Promoted ${configs.length} feature flags to production`);
}
```

### Step 6: Environment Indicator in UI

```typescript
// Show environment badge in development/staging
app.use((req, res, next) => {
  if (!IS_PROD) {
    res.setHeader('X-Environment', ENV);
  }
  next();
});

app.get('/api/status', (req, res) => {
  res.json({
    environment: ENV,
    version: process.env.npm_package_version,
    repl: process.env.REPL_SLUG,
  });
});
```

## Error Handling

| Issue | Cause | Solution |
|-------|-------|----------|
| Wrong database in prod | Manual DATABASE_URL override | Let Replit manage DB routing |
| Staging secrets in prod | Copied secrets incorrectly | Each Repl has independent Secrets |
| GitHub sync conflict | Both Repls on same branch | Use separate branches per Repl |
| Config validation fails | Missing env-specific secret | Add secret to correct Repl's Secrets tab |

## Resources

- Replit Deployments
- [Replit Secrets](https://docs.replit.com/replit-workspace/workspace-features/secrets)
- [Deploying from GitHub](https://docs.replit.com/hosting/deployments/deploying-a-github-repository)
- [PostgreSQL Dev/Prod Databases](https://docs.replit.com/cloud-services/storage-and-databases/postgresql-on-replit)

## Next Steps

For monitoring, see `replit-observability`. For deployment, see `replit-deploy-integration`.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.