bugiyz

Screenstudio MCP — AI skill for Claude Code

AI community

Screen Studio MCP Edit Screen Studio with Claude, gives AI agents tools to inspect, record, cut, zoom, preview, verify, and edit Screen Studio projects using natural language.

How to install Screenstudio MCP

This entry records only its repository, not the path inside it, so there is no exact command to give. Open bugiyz/screenstudio-mcp and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Screenstudio MCP does

Screen Studio MCP Edit Screen Studio with Claude, gives AI agents tools to inspect, record, cut, zoom, preview, verify, and edit Screen Studio projects using natural language. Instead of manually adjusting a product demo: “Make this feel like an Apple product video. Give the responses room, follow the conversation, verify the result.”

Alternatives in AI

  • Kimi Cc — Use Kimi's latest model (kimi-k2-0711-preview) to drive Claude Code 1.7k ★
  • Nodeterm — Node-based terminal manager for AI coding agents — tmux-backed terminals and parallel agent sessions as dragga 1.3k ★
  • Dinotty — Multi-device terminal server for AI coding agents 643 ★

README

Screen Studio MCP

Control and edit **Screen Studio** with Claude, Cursor, and other MCP-compatible AI clients.

`screenstudio-mcp` exposes Screen Studio automation as MCP tools by using the open-source [`screenstudio-agent`](https://github.com/HyperfocuSam/screenstudio-agent) CLI underneath.

Architecture

Claude / Cursor / MCP Client
            │
            │ MCP (stdio)
            ▼
     screenstudio-mcp
            │
            │ subprocess
            ▼
   screenstudio-agent
            │
            │ localhost Electron/CDP bridge
            ▼
       Screen Studio

What v0.1 can do

  • Launch Screen Studio automation
  • Check Screen Studio connection/status
  • Quit Screen Studio and close the automation bridge
  • List displays
  • List windows
  • Start recordings
  • Pause recordings
  • Resume recordings
  • Stop recordings
  • Cancel recordings
  • Verify the resulting .screenstudio project after recording
  • Inspect project information
  • Save projects
  • Undo and redo project changes
  • Seek to specific timestamps
  • Capture preview frames
  • Analyze recordings into frames and event timelines
  • Read Screen Studio configuration values
  • Modify Screen Studio configuration values
  • Trigger Screen Studio export
  • Quick-export to clipboard

Requirements

You need:

  • macOS
  • Screen Studio
  • Node.js 21 or newer
  • screenstudio-agent
  • ffmpeg and ffprobe for render/analyze features
  • macOS Accessibility permission for the terminal or MCP host controlling Screen Studio

Installation

There are two pieces:

  1. screenstudio-agent — talks to Screen Studio
  2. screenstudio-mcp — exposes those capabilities to AI agents through MCP

1. Install screenstudio-agent

Clone the underlying Screen Studio automation CLI:

cd ~
git clone https://github.com/HyperfocuSam/screenstudio-agent.git
cd screenstudio-agent
npm install
npx tsc

Create the `screenstudio` command:

mkdir -p ~/.local/bin
ln -sf "$PWD/bin/screenstudio.js" ~/.local/b