Sequencer

What the Sequencer MCP server does

The Sequencer MCP server gives an AI assistant access to media creation and project-editing workflows in Sequencer. It supports image, video, and speech generation, along with upfront cost estimates so a user can review planned credit usage before approving work.

The service also exposes connected workspaces, projects, and media. An assistant can assemble scenes and shots, adjust timelines, manage audio, add overlays or transitions, review creative output, and export finished projects. These capabilities place it in the 🎥 Multimedia Process category and make it relevant for assistants handling production tasks rather than only one-off asset generation.

How it works

Sequencer is provided as a hosted remote MCP endpoint at https://mcp.sequencer.media/. The connection uses Streamable HTTP and OAuth. After the client connects, the user signs in through a browser and approves access to the Sequencer account and its projects.

Tools and models are discovered from the hosted service, so the available operations are not fixed in the repository documentation. For generation tasks, the assistant should select an available model, present the complete settings, request a cost quote, and stay within the user-approved spending limit. Generated media and exports may take time to finish; the resulting links should be used only after the corresponding job completes.

Setup and configuration

Add a remote HTTP MCP server named sequencer in the assistant’s MCP settings and use the hosted endpoint as its URL. Start the connection, complete the Sequencer OAuth sign-in in the browser, and enable the tools in the assistant after authorization.

In GitHub Copilot for VS Code, use MCP: Add Server, or add an HTTP server entry to the workspace .mcp.json configuration. Credentials and access tokens should remain in the client’s secure authorization flow rather than being placed in the project configuration.

Tools and capabilities

The Sequencer MCP server can support workflows such as:

• Discovering available media models.
• Estimating generation costs before credits are spent.
• Generating images, videos, and speech.
• Checking generation status and retrieving completed media.
• Listing workspaces, projects, and connected media.
• Building scenes and shots and editing timelines.
• Managing audio, overlays, and transitions.
• Reviewing work and exporting completed projects.

The exact tool set depends on what the hosted service makes available to the connected account.

Limitations and notes

Some operations consume Sequencer credits and require user approval plus sufficient account balance. A client must support remote MCP connections and OAuth; this is not presented as a locally run server. Authorization grants the assistant access to the connected Sequencer account and its projects, so users should review the requested access before approving it. The repository contains connection documentation and a registry manifest, while the media-generation and editing operations are provided by the hosted service.