# How to Vibe Code Your Own Zernio (and Stop Paying for It) > Unified social media, messaging, and advertising API for developers and AI agents - Site: https://zernio.com - Category: Developer API & SaaS - Platforms: Web app, CLI, API - Verdict: **Don't bother** (18/100 vibecodeable) - Estimated effort: 6+ months of full-time work ## Verdict Keep paying for Zernio; building a multi-platform social API aggregator with 15 channels, telephony compliance, and an MCP server is an immense maintenance trap. Zernio wraps over 15 social graphs, 7 ad networks, WhatsApp WABA approvals, and telephony provisioning behind a clean API and MCP server. While you could vibecode a thin mockup for 2 platforms in a weekend, maintaining real API credentials, breaking changes across social networks, rate limiting, and webhook ingestions for 15 channels makes building this yourself an endless engineering sinkhole. The real value is not the UI dashboard; it is the tireless integration maintenance team dealing with upstream API deprecations daily. ### What you can't replicate - Upstream stability and ongoing maintenance across 15 constantly changing social APIs - Direct Meta WABA partnership compliance and global phone number KYC provisioning ## What it does Zernio is an API-first platform allowing developers and AI agents to publish, schedule, manage DMs, track analytics, run ad campaigns, and provision phone lines across 15+ social channels and 7 ad networks. ### Core features - Unified OAuth management for 15+ social channels - Cross-platform post publishing & scheduling queue - Unified social inbox for DMs, comments, and reviews - WhatsApp Business API and telephony (phone numbers, SMS, calls) - Social ads campaign management across 7 ad networks - Native Model Context Protocol (MCP) server exposing tools to AI agents ## The business ### Pricing - Self-Serve Usage-Based: Usage-based - Enterprise: Custom Founded 2025. Team size: 5. ## The hard parts - Maintaining 15+ disparate third-party social API integrations and authentication tokens - Strict anti-bot compliance, IP rotation, and avoiding account shadowbans - Resilient background workers and retry queues for media processing and publishing - Complex telecom and Meta regulatory compliance for phone numbers and WhatsApp Business API - Building and maintaining an MCP server with hundreds of deterministic AI tools ## How to vibe code Zernio ### Prerequisites - Node.js (free): Runtime environment for TypeScript backend services and Next.js dashboard. - GitHub (free): Source control and CI/CD triggers for Vercel deployment. ### Recommended AI tools - Claude Code: Best-in-class terminal coding agent for scaffolding multi-file integrations and handling complex API client logic. ### Stack - Frontend: Next.js - Backend: Next.js API Routes - Database: Turso - Auth: better-auth - Payments: Stripe - Other: Vercel AI SDK, Upstash, Resend ### Hosting - Vercel (Hosting the Next.js dashboard, API routes, and MCP server endpoints.): $0-20/mo - Turso (Serverless SQLite database for storing connected accounts, post queues, and webhook states.): $0/mo ### Build guide 1. **Project Scaffolding and Database Schema** — Initialize the Next.js project with TypeScript, configure Turso SQLite connection, and define core tables for users, connected social accounts, scheduled posts, and webhook event logs. ``` Create a new Next.js project with TypeScript and Tailwind CSS. Configure Drizzle ORM to connect to a Turso SQLite database over HTTP. Write the schema for 'users', 'accounts' (storing platform, access tokens, refresh tokens, and metadata), 'posts' (storing content, media URLs, target accounts, status, and scheduled publish times), and 'webhook_logs'. Implement database migration scripts and verify connection locally. ``` 2. **Unified OAuth Connect Framework** — Build dynamic API routes to handle OAuth authentication flows for Twitter/X, LinkedIn, and Instagram without requiring users to set up custom developer apps. ``` Implement an OAuth manager module in TypeScript that handles authorization code exchange and token refreshing for Twitter/X, LinkedIn, and Instagram. Create API routes GET /api/connect/{platform} that redirect users to the official social provider OAuth screens, and GET /api/connect/{platform}/callback that securely encrypts and stores access tokens in the Turso database linked to the user account. ``` 3. **Cross-Platform Publishing and Background Queue** — Develop a multi-platform publishing engine and asynchronous worker queue to deliver text and media posts across connected social accounts. ``` Build a robust posting service in TypeScript that accepts a payload with content, media items, and target account IDs. Implement individual adapter functions for Twitter/X and LinkedIn publishing APIs. Integrate Upstash QStash or background cron jobs to process scheduled posts from the 'posts' table, handling rate-limiting retries and logging failure reasons into webhook events. ``` 4. **Unified Social Inbox and Webhook Ingestion** — Create unified inbox endpoints to fetch and reply to DMs and comments across integrated channels, supported by incoming webhook processors. ``` Create API endpoints GET /api/inbox/conversations and POST /api/inbox/reply to aggregate incoming direct messages and comments from connected platforms. Implement webhook receiver endpoints POST /api/webhooks/{platform} to process real-time incoming events, normalize payloads into a unified format, and store them in the database. ``` 5. **Native Model Context Protocol (MCP) Server** — Build and expose a native Model Context Protocol (MCP) server endpoint containing deterministic tools for AI agents to interact with the social dashboard. ``` Implement a Model Context Protocol (MCP) server endpoint at /api/mcp using the Model Context Protocol SDK. Define tools for 'get_accounts', 'create_post', 'get_analytics', and 'send_message'. Ensure each tool executes database operations securely using the user's API key and returns structured JSON responses suitable for AI assistant consumption. ``` ### Cost vs paying **Starting costs (one-time):** - Domain name: $12 one-time - Developer API access fees (Twitter/X): $50 credit - Total: ~$62 one-time **Ongoing costs (monthly):** - Vercel Hobby/Pro hosting: $0-20/mo - Turso & Upstash usage: $0/mo - Total: ~$0-20/mo - Paying for the SaaS instead: Usage-based (~$24/mo for 4 accounts) - Build time: 160+ hours - AI tool credits: $20 (Claude Pro) - Break-even: N/A (Build is for learning only; Zernio is cheaper) ## Sources - [Zernio Website & Documentation](https://zernio.com) - [Zernio Pricing & Rate Card](https://zernio.com/pricing)