The official Model Context Protocol (MCP) server for VidPulse AI. Generate 16:9 YouTube videos, vertical Shorts/Reels, neural voiceovers, and animated subtitles directly inside your AI workflow.
Select your AI environment below to see step-by-step connection instructions.
In Claude.ai or the Claude App, go to Settings → Connectors → Add Custom Connector.
Name: VidPulse
URL: https://mcp.vidpulseai.site/mcp
Select Always Required (OAuth) or None. Click Connect to authorize your VidPulse account credits.
Add this configuration to your claude_desktop_config.json file:
{
"mcpServers": {
"vidpulse": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.vidpulseai.site/sse"],
"env": {
"VIDPULSE_API_TOKEN": "YOUR_VIDPULSE_JWT_TOKEN"
}
}
}
}
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Restart Claude Desktop after saving the configuration.
Add to Cursor Settings → Features → MCP Servers:
{
"name": "vidpulse",
"type": "streamable-http",
"url": "https://mcp.vidpulseai.site/mcp",
"headers": {
"Authorization": "Bearer YOUR_VIDPULSE_TOKEN"
}
}
Your IDE coding agent will be able to programmatically create AI videos, check job status, download generated media files, and verify video scene subtitles.
import requests
headers = {
"Authorization": "Bearer YOUR_VIDPULSE_TOKEN",
"Content-Type": "application/json"
}
payload = {
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "create_video_job",
"arguments": {
"topic": "Space Exploration 2026",
"aspect_ratio": "16:9",
"voice_id": "en-US-RogerNeural",
"caption_template": "mrbeast"
}
}
}
response = requests.post("https://mcp.vidpulseai.site/mcp", json=payload, headers=headers)
print(response.json())
Exposed natively to AI agents with complete type schemas and safety annotations.