premiere-pro

This Premiere Pro MCP Server exposes a set of MCP API tools that allow external automation and scripting to query and control Premiere Pro projects. You can retrieve project and sequence information, inspect timelines and clips, manage exports and renders, and perform sequence creation or trim operations on frames. Typical workflows include querying get_project_info to verify the active project, using get_active_sequence_info and get_sequence_details to drive an automation script, and calling trim_clip_by_frames to adjust clip in/out points via an external tool or UI integration. The server is designed to work with MCP clients and can be connected from AI IDEs, automation scripts, or Claude Desktop-like tools to orchestrate Premiere Pro tasks in a repeatable way.

Prerequisites:

• Adobe Premiere Pro (or compatible Adobe app)
• Node.js (LTS recommended) and npm
• Basic access to the Premiere MCP Server code repository (cloneable)

Install and run:

1. Clone the repository git clone https://github.com/yourusername/premiere-mcp-server.git cd premiere-mcp-server

2. Install dependencies npm install

3. Run the MCP server node mcp-server.js

4. Verify the server is running Open a browser or MCP client and connect to http://localhost:PORT (default port is 5000 unless you configure another port)

5. Optional: configure the CEP extension path and Premiere Pro integration as described in the README

Notes:

• If you modify the server, restart it to apply changes.
• Ensure firewall rules allow access to the configured PORT for MCP clients.

Tips and common issues:

• If you see Connection Refused or Status: Error: Failed to fetch, ensure the MCP server is running on the expected port and accessible from the host running the client.
• CEP extension compatibility may require enabling Developer Mode in Premiere Pro and placing the extension in the correct CEP extensions directory.
• Environment variables: PORT can be adjusted to avoid conflicts with other services. Document any required auth or restricted endpoints in your integration.
• When using get_export_presets or get_render_queue, ensure Premiere Pro is open with a project loaded; otherwise these calls may return empty results.
• For production deployments, consider adding authentication or IP whitelisting at the MCP layer to restrict access to trusted clients.