English Français العربية

Chat33

Official registry Marketing and sales

Publish to 8 social networks, answer Instagram and Facebook DMs, build comment-to-DM flows.

Source code Website Documentation

Install

Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).

Remote (streamable-http)
Claude Code 
claude mcp add --transport http chat33 'https://chat33.io/api/mcp/mcp'
JSON configuration (Claude Desktop, Cursor, VS Code) 
{
    "mcpServers": {
        "chat33": {
            "type": "http",
            "url": "https://chat33.io/api/mcp/mcp"
        }
    }
}

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

Contents
  1. Chat33 MCP Server
  2. Connect your client
  3. Tools
  4. Try asking
  5. Limits
  6. Links
  7. License

Chat33 MCP Server

Social media MCP server for AI agents. Connect Claude, ChatGPT, Cursor, VS Code or any MCP client to Chat33 and let your agent publish and schedule posts to Instagram, Facebook, Threads, TikTok, YouTube, X, LinkedIn and Bluesky, answer Instagram and Facebook DMs and comments, and build comment-to-DM automations.

This is a hosted (remote) server. There is nothing to install or run: point your client at the URL below and sign in.

https://chat33.io/api/mcp/mcp
  • Transport: Streamable HTTP
  • Auth: OAuth 2.1 sign-in in the browser (dynamic client registration), or an API key header
  • Docs: chat33.io/mcp

Connect your client

Prefer OAuth wherever your client supports it: add the server, choose Authenticate (or the client's login button) and approve in the browser tab that opens. No API key needed.

Claude Code
claude mcp add --transport http --scope user chat33 https://chat33.io/api/mcp/mcp

Then run /mcp, choose chat33 → Authenticate and approve in the browser.

Claude (claude.ai and Claude Desktop)

Settings → Connectors → Add custom connector → URL https://chat33.io/api/mcp/mcp → Connect, then approve in the browser.

ChatGPT

Settings → Apps / Connectors → Advanced → turn on Developer mode → Create: name Chat33, URL https://chat33.io/api/mcp/mcp, authentication OAuth. Sign in to Chat33 and click Allow.

Cursor

~/.cursor/mcp.json:

{
  "mcpServers": {
    "chat33": {
      "url": "https://chat33.io/api/mcp/mcp"
    }
  }
}
VS Code

.vscode/mcp.json in your workspace (or the user MCP config):

{
  "servers": {
    "chat33": {
      "type": "http",
      "url": "https://chat33.io/api/mcp/mcp"
    }
  }
}
Windsurf

~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "chat33": {
      "serverUrl": "https://chat33.io/api/mcp/mcp"
    }
  }
}
Codex
codex mcp add chat33 --url https://chat33.io/api/mcp/mcp
codex mcp login chat33
Gemini CLI
gemini extensions install https://github.com/chat33-io/chat33-mcp
Clients without OAuth (API key)

Create a key in Chat33 → Settings → API Keys and send it as an x-api-key header (or Authorization: Bearer):

{
  "mcpServers": {
    "chat33": {
      "url": "https://chat33.io/api/mcp/mcp",
      "headers": { "x-api-key": "sk_live_..." }
    }
  }
}

Tools

GroupTools
Accountlist_accounts, whoami
Publishingcreate_post, validate_post, list_posts, get_post, delete_post, get_post_logs, request_approval, list_approvals, add_post_note
Mediaupload_media, create_upload_url
Analyticsget_analytics
Automationslist_flows, get_flow, create_flow, set_flow_status
Inboxlist_conversations, get_messages, reply_to_conversation
Commentslist_post_comments, reply_to_comment

Tools that message a real person (reply_to_conversation, reply_to_comment) tell the agent to ask for your approval first. New automations are created as drafts.

Try asking

  • "Schedule this photo to Instagram and Threads for tomorrow at 9am."
  • "Which DMs are waiting for a reply? Draft answers for me to approve."
  • "Build a flow that DMs the guide link to anyone who comments GUIDE on my last reel."
  • "How did my posts do this week?"

Limits

  • Your Chat33 plan's limits on posts and connected accounts apply.
  • Instagram and Facebook DM replies only work within 24 hours of the contact's last message.
  • Meta allows one private reply per comment, within 7 days.
  • LinkedIn: personal profiles only. Bluesky: publishing only (no DMs or comments).

License

MIT. This repository describes the hosted server; the server itself runs at chat33.io.

Frequently asked questions

What is Chat33?

Publish to 8 social networks, answer Instagram and Facebook DMs, build comment-to-DM flows.

How do I install Chat33?

It runs as a remote server. In Claude Code: claude mcp add --transport http chat33 'https://chat33.io/api/mcp/mcp'. For other clients, paste the JSON configuration from this page into their MCP settings.

Does Chat33 need an API key?

Its registry entry lists no required key or secret. Some features may still need an account with the service it connects to.

Is Chat33 open source?

Yes. Its code is on GitHub under the MIT license.

Who maintains Chat33?

It is published by chat33-io. AI.tn lists it but does not maintain or endorse it.

From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.