Zernio MCP Server

The Zernio MCP Server implements the Model Context Protocol on top of Zernio's own API and opens up social media and messaging accounts to external AI systems. Instead of building a separate OAuth integration and API client for every platform, an MCP client connects to exactly one endpoint and gets access to all connected accounts through it. Technical documentation lives at docs.zernio.com/mcp; the service is operated by the Zernio organization (GitHub org zernio-dev).

Tool architecture

Per official documentation, a core group of 20 tools covers the most common flows, including creating, scheduling, and drafting posts, managing connected accounts, retrieving analytics and engagement data, creating ad campaigns, controlling the publishing queue, and validating and checking the health of connected accounts. Beyond that, the server automatically generates one tool per endpoint of its OpenAPI specification. To keep this large tool catalog from overwhelming every conversation's context, additional tools can be found via search_tools and loaded on demand via call_tool, rather than being fully present in the language model's context from the start.

Supported platforms and ad networks

According to the homepage, the server covers 16 social and messaging platforms, including Instagram, TikTok, YouTube, X/Twitter, LinkedIn, Facebook, Threads, Pinterest, Reddit, Bluesky, WhatsApp, Telegram, Discord, Snapchat, and Google Business Profile. In addition, ad campaigns can be managed on seven networks: Meta Ads, Google Ads, TikTok Ads, LinkedIn Ads, Pinterest Ads, X Ads, and OpenAI Ads (ChatGPT Ads). This breadth makes the server particularly interesting for teams running multiple channels in parallel that previously had to maintain several separate integrations.

Setup and authentication

The server is available exclusively as a hosted remote service at https://mcp.zernio.com/mcp via Streamable HTTP; per the documentation, there is no local installation path or open-source repository for the server code itself. Two authentication paths are available: MCP clients with OAuth support use OAuth 2.1 with PKCE and need only the server URL; clients without OAuth support and autonomous agents instead authenticate via an API key sent as an Authorization: Bearer header. For Claude Code, the documentation lists two setup commands: claude mcp add --transport http zernio https://mcp.zernio.com/mcp followed by claude /mcp to sign in. Per the provider, Cursor, Codex, ChatGPT, Windsurf, VS Code, and other standard MCP clients are also supported.

Open-source companion projects on the same API

In the zernio-dev GitHub organization, Zernio publishes several open-source tools built on the same underlying API, including the social media scheduler Latewiz, the visual chatbot builder Zernflow, a cross-platform unified inbox, an ads reporting dashboard, and an official Node.js SDK and CLI tool. These projects are distinct from the hosted MCP server described here: the MCP server itself remains a proprietary, hosted-only service with no public source code of its own.

Pricing model

Per the provider, the first two connected social media accounts are permanently free to use, including a monthly inbox message allowance, unlimited posts, analytics, and full API access. Beyond that, a usage-based, tiered pricing model applies per additional connected account, where the per-account rate decreases as more accounts are connected; metered costs also apply for inbox messages beyond the free allowance, for managed ad spend beyond a free allowance, and optionally for phone numbers, calls, SMS, and WhatsApp usage. No fixed plans or seat-based pricing, and no credit card is required for the free tier. Current amounts are available at zernio.com/pricing.

Security and write access

The tool catalog includes far-reaching write permissions: posts can be published and scheduled, inbox messages replied to, and ad campaigns with real budgets created, across multiple platforms simultaneously. This reach calls for deliberate permission handling: API keys should be stored securely like any other secret, OAuth consent should be granted only for accounts actually needed, and automated ad spending should be capped before an agent launches campaigns on its own. Since inbox content and comments from external users can feed into the connected language model's responses, agent output should be reviewed before any production publishing.

Who the server is a good fit for

The Zernio MCP Server is particularly useful for marketing teams, agencies, and solopreneurs running multiple social media and messaging channels in parallel who want to use an AI agent instead of juggling individual platform dashboards. Compared to single-platform servers such as the already-listed X API MCP Server, Zernio covers not just one but 16 social channels plus ad networks through the same endpoint, making it best suited for cross-channel automation rather than the depth of a single platform's API.

A Zernio account with at least one connected social media or messaging platform, an MCP client with streamable HTTP/remote support, and either OAuth 2.1 consent or a Zernio API key.

In Claude Code, two commands are enough:

claude mcp add --transport http zernio https://mcp.zernio.com/mcp
claude /mcp

The second command opens sign-in; for OAuth-capable clients, browser consent is all that's needed afterward. In clients without native OAuth support, or for autonomous agents, a Zernio API key is set as an Authorization: Bearer header instead.

claude mcp add --transport http zernio https://mcp.zernio.com/mcp

OAuth 2.1 with PKCE for clients with native OAuth support (only the server URL is needed), or a Zernio API key sent as an Authorization: Bearer header for clients without OAuth support and for autonomous agents.

Access scope depends on the connected Zernio account: via OAuth or API key, the client gets access to all social media and messaging platforms connected to that account, along with their posts, inboxes, analytics, and ad campaigns.

Requests go through the Zernio-hosted endpoint mcp.zernio.com via Streamable HTTP. Responses can include post, inbox, analytics, and ad campaign data from connected social media accounts and are returned to the MCP client and its language model.

The server offers far-reaching write access: posts can be published and scheduled, inbox messages replied to, and ad campaigns with real budgets created across multiple platforms. API keys should be stored securely like any other secret, OAuth consent should be limited to accounts actually needed, and ad spending should be capped. Manipulated inbox or comment content from external users could theoretically carry prompt injection attempts against the connected AI model; review agent output before production publishing and campaign launches.