fcpxml
🎬 The first AI-powered MCP server for Final Cut Pro XML. Control your edits with natural language.
claude mcp add --transport stdio daredev256-fcpxml-mcp-server python server.py \ --env FCP_PROJECTS_DIR="Path to your Final Cut Pro projects directory"
How to use
This MCP server, named fcpxml, acts as a bridge between Final Cut Pro XML exports and Claude AI workflows. It exposes a suite of tools (53 operations referenced in the project) that analyze, modify, and generate data from FCPX XML timelines. Typical usage involves exporting an XML from Final Cut Pro, sending it to Claude Desktop or your MCP client to run analyses, then applying non-destructive edits or generating structured data (markers, chapters, metadata) which can be re-imported or used for downstream tasks. The server is designed to keep original media untouched while producing structured, actionable data that Claude can reason about and transform according to prompts such as health checks, marker generation, or pacing analyses. Tools cover extraction (clips, markers, keywords), validation (health scores, gaps, flash frames), and generation (rough cuts or chapter markers) to accelerate editing pipelines and QA workflows.
To use it, configure Claude Desktop or your MCP client to point at the fcpxml server (as configured in mcp_config), ensure Python 3.10+ is installed, and run the server script. Then you can invoke the available MCP tools to analyze your FCXP timeline XML, request fixes or augmentations, and export results in formats useful for reporting or re-import back into your NLE workflow.
How to install
Prerequisites:
- Python 3.10 or newer installed on your system
- Git
- Access to a terminal/command prompt
Step-by-step:
-
Clone the repository git clone https://github.com/DareDev256/fcpxml-mcp-server.git cd fcpxml-mcp-server
-
Install the package in editable mode (recommended during development) python -m pip install --upgrade pip pip install -e .
-
Ensure you have a directory for your Final Cut Pro projects and set the environment variable as needed (see mcp_config) Example (bash): export FCP_PROJECTS_DIR="/Users/you/Movies"
-
Run the server script (as configured in mcp_config) python server.py
-
Configure Claude Desktop or your MCP client to connect to the fcpxml MCP server using the provided mcp_config settings.
Additional notes
Notes and tips:
- The server expects a valid FCP XML export as input for most tools; results are designed to be non-destructive to your original XML.
- The example environment variable FCP_PROJECTS_DIR should point to where you store FCP projects; tools may reference this path to locate assets or additional data.
- If you encounter path or permission errors, verify that the Python process has read/write access to your project directories and that the server.py file is executable in your environment.
- When using Claude Desktop, you can place fcpxml under Claude’s mcpServers in your local configuration (as shown in the readme) to streamline tool invocation.
- If you upgrade Python or dependencies, re-run pip install -e . to ensure the editable install stays up to date.
- This MCP server is designed to work with XML exports from Final Cut Pro 10.4+ as indicated by the repository’s scope and requirements.
Related MCP Servers
mcp-telegram
MCP Server for Telegram
lc2mcp
Convert LangChain tools to FastMCP tools
boilerplate
TypeScript Model Context Protocol (MCP) server boilerplate providing IP lookup tools/resources. Includes CLI support and extensible structure for connecting AI systems (LLMs) to external data sources like ip-api.com. Ideal template for creating new MCP integrations via Node.js.
docmole
Dig through any documentation with AI - MCP server for Claude, Cursor, and other AI assistants
skill-to
Convert AI Skills (Claude Skills format) to MCP server resources - Part of BioContextAI
whatsapp -extended
Extended WhatsApp MCP server with 41 tools - reactions, group management, polls, presence, newsletters & more. Fork of lharries/whatsapp-mcp