Connect Shotstack to AI Tools via MCP for Video Generation

Description and use

Integrating Shotstack with AI assistants via MCP enables video creation and editing through chat without writing JSON timelines. Supports Claude Desktop and Cursor.

Problem

Creating videos with AI assistants typically requires manually writing complex JSON timelines or making direct API calls, which is tedious and error-prone for non-developers.

Outcome

Users can generate and edit videos directly through AI chat interfaces by connecting Shotstack via MCP. The workflow supports both direct rendering to URL and interactive editing in Studio, eliminating the need for manual JSON timeline construction.

Workflow

• Obtain a Shotstack API key (sandbox or production) from app.shotstack.io.
• For Claude Desktop: Go to Customize > Connectors, add a custom connector with URL https://mcp.shotstack.io/, and complete OAuth.
• For Cursor: Add the MCP server configuration to ~/.cursor/mcp.json with the URL https://mcp.shotstack.io/.
• Prompt the AI assistant to create a video (e.g., 'Create a 15-second video about the earth...').
• The AI calls the 'render_video' tool, which returns a render ID.
• The AI polls 'get_render_status' until the video is done.
• The AI provides the output URL or offers to open the 'studio' tool for interactive editing.
• Review and download the final video from the Shotstack dashboard or editor.

Evidence

Shotstack ships an official MCP server. That means you can connect it directly to Claude Desktop, Claude Code, Cursor, ChatGPT, Antigravity, or any other MCP-aware client, and create or render videos by chatting — no hand-written JSON timelines, no manual API calls.

Source

Enter the URL: https://mcp.shotstack.io/ ... Click on the Connect button and then complete the OAuth flow

Source