videoFn

io.github.stevejhiggsPublished by GitHub account stevejhiggsv1.4.4Updated Nov 1, 2026

Video editor for your AI assistant. Builds, checks and renders edits on your Mac.

VerifiedSTDIODesktop onlyFiles & Storage Media & Design

Server details

Identifier: io.github.stevejhiggs/videofn
Transport: STDIO
Runtime: Desktop only
Trust: Verified
Published: Nov 1, 2026

Security note

MCP servers receive tool arguments from a conversation and may perform external actions. Review what a server can do before you enable it, and give it only the access it needs.

The registry verified the GitHub account stevejhiggs for this listing. That does not show the account is the product's maker.

SourceWeft's checks cover whether the listing is complete and the endpoint answered; they do not review the server's code or what its tools do.

Add a badge to your README

Show readers this MCP server is listed on SourceWeft. The badge links to this page.

[![Listed on SourceWeft](https://sourceweft.com/badge/mcp/io.github.stevejhiggs%2Fvideofn.svg)](https://sourceweft.com/mcp/io.github.stevejhiggs%2Fvideofn?utm_source=readme&utm_medium=badge)

Overview

AI-generated overview

videoFn lets an AI assistant build, check and render video edits on an Apple Silicon Mac through the native videoFn editor app.

What you can do: Build edits from footage, screen recordings, photos, music or documents
Check the edit for text legibility, cut pacing, speech cuts, audio/visual sync and the exported file, then fix issues
Show a live preview and accept notes in plain English
Transcribe, generate voiceover, reduce noise, track subjects and remove image backgrounds on the device
Export the finished MP4
When to use it: Useful when you edit video on an Apple Silicon Mac and want an assistant to assemble cuts, captions, music and explainers from your own files, or to clean up talking-head recordings, while you review a live preview.
Requirements: An Apple Silicon Mac with the videoFn app installed, plus an MCP client such as Claude Desktop, Claude Code, ChatGPT or Codex. The Claude Desktop connector is a small launcher that starts the broker inside the installed app. No account or API keys are declared.

Before you install

Free to try with a watermark on exports; removing it requires a one-time purchase, and your AI assistant is billed separately by its provider. Your assistant receives transcripts, measurements and requested preview frames under its provider's terms. The connector is desktop-only and needs the app installed.

Installation

In SourceWeft

Open videoFn in the dashboard and add it to a workspace.
Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

videoFn MCP

Video editor for your AI assistant. Builds, checks and renders edits on your Mac.

videoFn gives Claude, ChatGPT, Codex or any MCP client the tools to edit video on your Mac. Your assistant builds the edit, checks its own work and fixes the problems it finds. You watch the live preview, give notes in plain English and export the finished MP4.

This repository publishes the videoFn connector for Claude Desktop (videoFn.mcpb) and the metadata for the MCP Registry. The videoFn app itself is at videofn.com.

What it does

videoFn is a native video editor for Apple Silicon Macs with a built-in MCP server. You brief your AI assistant, and it uses videoFn's tools to build the edit from your footage, screen recordings, photos, music or documents.

It checks its own work. videoFn measures text legibility against moving backgrounds, cut pacing, speech cuts, audio and visual sync, and the exported file. The assistant fixes what the checks find.
You stay in control. The live preview shows the edit as it takes shape. Give notes in plain English, or change timing yourself in the editor.
Rendering happens on your Mac. No uploads, no render queue. Transcription, voiceover, noise reduction, subject tracking and image background removal also run on the device. Your assistant receives transcripts, measurements and the preview frames it asks for, under its provider's terms.
Buy it once. videoFn is free to try with a watermark on exports. One purchase removes the watermark. No account, no subscription, no credits. Your AI assistant is billed separately by its provider.

What people make with it

Product demos and launch videos from screen recordings
App Store preview videos
Release notes and changelog videos
Vertical social clips with captions from longer recordings
Explainers from a script, a PDF, a Word document or a spreadsheet
Talking-head cleanup: remove filler words and long pauses, add subtitles
Beat-synced edits, lyric videos, music videos from photos, YouTube intros

Example prompts

"Make a 30-second launch video from the screen recording in my Downloads folder. Add captions and some upbeat music."
"Cut the ums and long pauses from this interview and add subtitles."
"Turn this PDF report into a two-minute narrated explainer with animated charts."
"Make a vertical clip of the best 45 seconds of this podcast recording."

Requirements

An Apple Silicon Mac
The videoFn app
An MCP client: Claude Desktop, Claude Code, ChatGPT, OpenAI Codex or another client

Install

Download videoFn from videofn.com/download and open it.
On the launch screen, select Connect for your assistant.
Ask your assistant for a video.

Claude Desktop only: you can also install the connector first. Download videoFn.mcpb from the latest release and open it. The connector needs the videoFn app. Until the app is installed, it tells Claude where to download it.

How the connector works

videoFn.mcpb contains a small launcher and no editor code. The launcher starts the MCP broker inside the installed videoFn app, so the connector always matches the app version you have. The broker starts the app when your assistant first needs it.

Licence

The files in this repository, including the launcher in videoFn.mcpb, are under the MIT licence. The videoFn app is proprietary software.

Source: README.md at commit d6aa30c

Tools

SourceWeft started this package once in an isolated sandbox, without credentials, to list its tools. That failed: the package type cannot run in the sandbox.

Version history

v1.4.4LatestNov 1, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
DescriptionRequired15 pts
Logo10 pts
Documentation15 pts
License10 pts