Skip to main content
Back

Higgsfield AI's Personal Clipper Turns Claude Chats Into a Video Editing Suite

Higgsfield's Personal Clipper brings AI-powered viral clip extraction directly into Claude via MCP, turning any long-form video into platform-ready shorts in one conversation.

Content repurposing has always been the unglamorous half of the creator workflow. You record or license a great long-form video, then spend hours in a timeline scrubbing for the moments worth clipping, resizing for every platform, burning in subtitles, and exporting five different versions. Higgsfield AI just moved that entire process into a Claude chat window.

What just shipped

Higgsfield's Personal Clipper is now available inside Claude through the Higgsfield MCP connector. Drop a video link into the chat, tell Claude how many clips you want and which moments to prioritize, pick an aspect ratio and subtitle font, and get back a set of ready-to-post clips. No timeline scrubbing, no export queue, no switching apps.

The workflow is four steps:

  1. Drop the link to your video into Claude
  2. Ask Higgsfield MCP to cut it into X clips based on the most viral moments
  3. Choose the aspect ratio and subtitle font
  4. Get the clips, ready to post across every social platform

The feature sits on top of the broader Higgsfield MCP connector, which has been expanding steadily since its launch. Higgsfield shipped its official MCP server on April 30, 2026, exposing 30+ image and video generation models through a single hosted endpoint. Personal Clipper is the first tool in that suite aimed squarely at editing existing footage rather than generating new content from scratch.

The plumbing underneath

MCP, short for Model Context Protocol, is an open standard that gives AI agents secure access to external tools. Claude supports it natively, and Higgsfield runs an MCP server at https://mcp.higgsfield.ai. Connecting the two gives Claude direct access to the entire Higgsfield platform from any chat, so your conversation thread effectively becomes your production environment.

Three components are doing the work behind the scenes when you run a clip request:

  • Claude interprets your intent, identifies the viral-potential moments, and turns natural language into structured generation parameters
  • MCP acts as the secure bridge, with your credentials staying on Higgsfield's side and Claude only ever seeing the tool results
  • Higgsfield runs the actual clip extraction, subtitle rendering, and aspect ratio conversion, then returns the finished assets back to chat

A typical prompt looks like: "Take this 18-min YouTube interview and cut 3 vertical clips for TikTok. Add subtitles in 'Bebas Neue'