Photoshop MCP Server
Enables AI clients to control Adobe Photoshop on macOS through AppleScript using the Model Context Protocol.
README
Photoshop MCP Server (AppleScript)
A Model Context Protocol (MCP) server that enables AI clients to control Adobe Photoshop on macOS through AppleScript. This project adapts the architecture from adb-mcp using native macOS automation instead of a Node.js proxy and UXP plugin.
Features
- Document Operations: List open documents, retrieve document metadata (dimensions, resolution, aspect ratio)
- Layer Management: List layers with hierarchy, rename layers, toggle visibility, adjust opacity and blend modes
- Layer Organization: Create new pixel layers, duplicate layers, delete layers, reorder layers (top/bottom/up/down)
- Document Control: Save documents, flatten all layers
- MCP Compatible: Works with any MCP-compatible AI client (Claude Desktop, OpenAI Agent SDK, etc.)
Requirements
- macOS with AppleScript support
- Adobe Photoshop 2026 or later
- Python 3.10 or later
- MCP-compatible client
Installation
1. Clone the repository
git clone https://github.com/<username>/photoshop-mcp.git
cd photoshop-mcp
2. Create virtual environment
python3 -m venv venv
source venv/bin/activate
3. Install dependencies
pip install mcp fastmcp
Configuration
Set the PHOTOSHOP_APPLICATION_NAME environment variable if your Photoshop installation has a different name:
export PHOTOSHOP_APPLICATION_NAME="Adobe Photoshop 2026"
Default value is Adobe Photoshop 2026.
Usage
Start the MCP server
python3 photoshop_mcp.py
The server communicates via stdio and exposes tools to MCP clients.
Available Tools
| Tool | Description |
|---|---|
get_documents |
Returns a list of open Photoshop documents with name, path, and modified status |
get_document_info |
Returns current document metadata: width, height, resolution, pixel aspect ratio |
get_layers |
Returns all layers with id, name, type, visibility, opacity, blend mode, and nested children |
set_layer_visibility |
Sets visibility of a layer by id |
set_layer_properties |
Updates opacity, blend mode, and visibility for a layer |
rename_layer |
Renames a layer by id |
create_pixel_layer |
Creates a new art/pixel layer |
duplicate_layer |
Duplicates a layer by id |
delete_layer |
Deletes a layer by id |
move_layer |
Moves a layer by id. Positions: TOP, BOTTOM, UP, DOWN |
get_layer_bounds |
Returns pixel bounds of a layer |
flatten_all_layers |
Flattens all layers into a single layer |
save_document |
Saves the current document |
Example Prompts
Once connected to an MCP client:
List all layers in the current Photoshop document.
Rename layer 3 to "Background - Final".
Set layer 5 to 50% opacity and multiply blend mode.
Create a new layer named "Adjustment Layer" with 80% opacity.
Architecture
AI Client <-> MCP Server (Python/FastMCP) <-> AppleScript <-> Adobe Photoshop
This implementation replaces the Node.js proxy server and UXP plugin architecture from the original adb-mcp with direct AppleScript execution, reducing complexity and external dependencies.
Project Structure
photoshop-mcp/
photoshop_applescript.py # AppleScript wrappers for Photoshop operations
photoshop_mcp.py # FastMCP server exposing tools
pyproject.toml # Project metadata and dependencies
README.md # Documentation
ROADMAP.md # Development roadmap
Development
Running tests
python3 -c "
import sys
sys.path.insert(0, '.')
from photoshop_applescript import get_documents, get_layers
print(get_documents())
print(get_layers())
"
Adding new tools
- Add an AppleScript wrapper function in
photoshop_applescript.py - Expose it as an MCP tool in
photoshop_mcp.pyusing the@mcp.tool()decorator
Roadmap
- [x] Core layer editing (rename, visibility, opacity, blend mode)
- [x] Layer creation, duplication, deletion, reordering
- [x] Document info and document listing
- [x] Layer bounds retrieval
- [x] Save and flatten operations
- [ ] Image export (PNG, JPG)
- [ ] Layer image extraction
- [ ] Place image into layer
- [ ] Selection tools
- [ ] Fill and delete selection
- [ ] Adjustment layers
- [ ] Text layers
- [ ] Layer masks
License
MIT
Acknowledgments
Inspired by adb-mcp by Mike Chambers.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.