Model Context Protocol (MCP) Server
Empower your AI agents with direct access to social media publishing via PostPulse.
Overview
The Model Context Protocol (MCP) allows AI assistants like Claude Desktop, Cursor, and other MCP-compatible clients to interact with external tools and data. By connecting the PostPulse MCP Server, your AI agent can list accounts, upload media, and schedule posts — all directly from your chat interface.
🤖 AI-Native
Works with Claude Desktop, Cursor, and any MCP-compatible AI client
🔐 Secure OAuth2
Full MCP OAuth2 auto-discovery — no API keys needed, just authorize once
📱 Multi-Platform
Publish to all 9 supported platforms from a single AI conversation
🧰 4 Powerful Tools
List accounts, get chats, upload media, and schedule posts
Setup
Connect the PostPulse MCP Server to your AI assistant.
Remote MCP Server (Recommended)
Connect in your AI client
Add the remote MCP server URL to your client:
https://mcp.post-pulse.comYour client will auto-discover the OAuth endpoints and walk you through authorization.
Claude Desktop
For Claude Desktop, add this to your claude_desktop_config.json:
{
"mcpServers": {
"postpulse": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://mcp.post-pulse.com"
]
}
}
}Available Tools
The PostPulse MCP Server exposes 4 tools that your AI agent can call.
list_accounts
List all connected social media accounts (Instagram, Facebook, Telegram, etc.) with their IDs and platforms.
Parameters:
None — returns all connected accounts.
Returns:
Array of account objects with id, platform, username, and name.
list_chats
List chat threads for a connected social media account. Required for Facebook Pages and Telegram Channels.
Parameters:
| Name | Type | Description |
|---|---|---|
accountId | number | The account ID from list_accounts |
platform | string | Platform name (e.g. INSTAGRAM, FACEBOOK, TELEGRAM) |
Returns:
Array of available chats (Telegram channels / Facebook pages) with IDs and titles.
upload_media
Upload media from a URL or binary data for use in scheduled posts. Returns the media path/key.
Parameters:
| Name | Type | Description |
|---|---|---|
mediaUrl | string (optional) | Public URL of the media file to upload |
mediaData | string (optional) | Base64-encoded media data |
mediaType | string (optional) | MIME type (required if mediaData is provided) |
mediaName | string (optional) | Name of the media file |
Returns:
The media path/key to use in schedule_post's mediaPaths parameter.
schedule_post
Schedule a social media post to one of your connected accounts. Supports Instagram, Facebook, YouTube, TikTok, Threads, LinkedIn, Telegram, X (Twitter), and Bluesky.
Parameters:
| Name | Type | Description |
|---|---|---|
accountId | number | The account ID from list_accounts |
platform | string | Platform name (INSTAGRAM, FACEBOOK, TELEGRAM, etc.) |
content | string (optional) | Post content text |
mediaPaths | string[] (optional) | Media paths from upload_media |
scheduledTime | string | ISO-8601 timestamp (e.g., 2025-10-27T10:00:00Z) |
publicationType | string (optional) | FEED, REEL, or STORY (Instagram/Facebook) |
title | string (optional) | Video title (YouTube/TikTok) |
facebookPageId | string (optional) | Facebook Page ID from list_chats |
telegramChannelId | string (optional) | Telegram Channel ID from list_chats |
topicTag | string (optional) | Topic tag for Threads |
Returns:
Confirmation message with the scheduled post ID.
Example Workflow
Here's a typical AI-driven workflow: ask your agent to create an image, upload it via PostPulse, and schedule a post — all in one conversation.
