Photoshop-MCP-IMT + SKILLS — Photoshop for AI agents, with two ways in

Let an AI agent (Claude Desktop, Cursor, ChatGPT or any MCP-compatible client) drive a running Adobe Photoshop: documents, layers, text, masks, selections, filters, adjustments, Smart Objects, artboards, generative tools, export — from natural language.

Every Adobe UI language · a large skills library · built for generative features. Photoshop-MCP-IMT works with every UI language Adobe ships, comes with a large skills library and is built for Adobe's generative features — Generative Fill, Generative Expand, Generative Upscale, text to image, Firefly and partner models, generative credits.

Two parts, made to work together:

• The MCP server — 135 tools behind a 10-tool facade, a second transport inside Photoshop for the moments a dialog blocks scripting, honest errors, no telemetry.
• The skill (skills/ps, also served by the ps_skill tool, so it works in any client) — how to drive Photoshop safely, where scripting ends, and how to use Photoshop's own AI tools: Generative Fill / Expand / Remove, Generate Image and Background, partner models and their credits, reference images, Prompt to edit, Harmonize, Generative Upscale, the Remove tool, AI Assistant — plus the retouch, colour, selection and batch recipes that combine them, and the gotchas.

Built by Immersive Media Technologies. A fork of alisaitteke/photoshop-mcp (merged through 1.7.27) with a second transport, a small tool surface, honest errors and no telemetry.

macOS and Windows · Photoshop 2024–2026 · Node.js 18+ (we run 26). macOS is what we run every day (Photoshop 2026, 27.9–27.10); the Windows transport (ExtendScript through COM, with the UTF-16 result fix and a retry on the COM «busy» error) passes the same live sweep on Windows 11 with Photoshop 2026 — reports from Windows users are still welcome.

[!CAUTION] This tool edits real Photoshop documents and sends document contents (names, layer structure, text, rendered previews) to the AI service you use. Start with a copy, check your AI provider's data policy for NDA work, and keep ps_execute_script and ps_uxp script (arbitrary code inside Photoshop) behind your agent's permission policy.

Why this one

Most Photoshop MCP servers stop at «the AI can run a script». Two things break that in daily use: Photoshop goes modal (a dialog, a progress bar, a tool in search mode) and every script call hangs until a timeout; and the agent drowns in a hundred tool definitions. This fork keeps a second transport inside Photoshop (UXP) for exactly those moments, shows the agent 10 tools instead of 135, and tells it the truth when something cannot be done.

Built for localized Photoshop, not just English. Most Photoshop MCP servers read Photoshop's English messages and names — on a German, Japanese or Russian install their error handling and layer lookups break. Photoshop-MCP-IMT classifies errors by Photoshop's error numbers and finds the Background layer, fonts and actions without localized names, so it behaves the same on every UI language Photoshop ships — verified live.

| | Photoshop-MCP-IMT (this repo) | alisaitteke/photoshop-mcp | mikechambers/adb-mcp | dcc-mcp/dcc-mcp-photoshop | | ----------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------- | -------------------------------------------------------- | | Talks to Photoshop via | ExtendScript (osascript / COM) and a UXP panel plugin — chosen per call, UXP as fallback or first (PS_MCP_UXP_PREFER=1) | ExtendScript via AppleScript / COM (we are a fork) | UXP plugin → separate WebSocket proxy (Python) → MCP server | UXP plugin → Rust broker (adobepy) → Python MCP server | | When Photoshop is modal | photoshop_busy in 4 s with an instruction for the user, 8 s hold; reads / save / export continue through UXP (via: uxp) | 30 s timeout → extendscript_timeout, retry ping | — | UXP only, so unaffected — but no ExtendScript DOM at all | | Tool surface | 135 tools registered, 10 visible (ps_catalog / ps_do / ps_batch …) — ~900 tokens of definitions per turn instead of ~25 K; every tool reachable through ps_do | 118–122 tools, all visible (photoshop_* + 16 recipes) | «a subset of functionality» per app, low-level tools (proof of concept) | 40+ typed tools in 8 skill packages | | Headless corner-pin | photoshop_distort_layer: four corners → custom warp solved for the homography; verified that Trnf + quadrilateral is silently ignored in Photoshop 2026 (12 combinations probed) | — | — | — | | Telemetry | none — analytics module is a no-op, feedback nudge removed, server sends nothing | anonymous analytics on by default (opt-out), product-feedback nudge via photoshop_ping | — | — | | Undo | recipes = one history state (upstream); photoshop_busy / command_unavailable / extendscript_timeout / scratch_disk_full tell the agent why instead of a generic timeout | recipes = one history state; extendscript_timeout, scratch_disk_full | — | staged install with --dry-run / receipts | | Install | one click: .mcpb for Claude Desktop (no Node needed), «Add to Cursor» button, npx for any other client; install.sh / install.ps1 with a live self-test that write the client configs; UXP plugin as .ccx, UXP Developer Tools or a manual layout — every path documented with its failure modes | npx -y @alisaitteke/photoshop-mcp; UXP plugin for Neural Filters only | Python 3 + Node + UXP Developer Tools + proxy process | pip install or binaries; Rust broker | | Verified on | Photoshop 2026 (27.9–27.10), macOS, UXP 9.4.1 — live sweep of every tool: 193 pass / 0 fail / 11 environment skips (2026-10-01); Windows 11 (ARM) + Photoshop 2026 — same sweep, 193 / 0 / 11 (2026-10-04) | Windows + macOS | Photoshop 26+, macOS + Windows | Windows, Linux, macOS | | UI language | every UI language Photoshop ships — errors classified by error numbers, not localized messages; the Background layer, fonts (PostScript name) and actions found without localized names; verified live | License | Apache-2.0 for our work (attribution through NOTICE); upstream code stays MIT | MIT | MIT | MIT |

Facts about other projects are from their READMEs and repository metadata on GitHub on 2026-10-01 (adb-mcp: 716 stars, last push 2026-07-08, «proof of concept»; dcc-mcp-photoshop: v0.2.0; alisaitteke/photoshop-mcp: 1.7.27 of 2026-09-30); corrections welcome.

What you can ask for

• "Open poster.psd, put the client's logo on the billboard in perspective — corners at (412,88), (1388,140), (1370,690), (430,655) — and keep it editable."
• "Replace the sky with this dusk one (sky.jpg), then export a 2000-px JPEG for review."
• "Split this layout into Instagram, X and Story artboards and export each as PNG."
• "Set the headline to Inter Bold 96 pt, tracking −20, only the word SALE in red."
• "Build a frequency-separation stack at 6 px — I'll paint the Low / High layers myself."
• "Photoshop is showing a dialog — tell me what is open and save the document anyway." (the save goes through the UXP bridge)

The agent combines the operations itself through ps_do, checking ps_get_state / ps_get_layers between steps and verifying with ps_export_frame (a PNG of the current state).

The skill

skills/ps/SKILL.md is a short router the agent reads first (ps_skill with no arguments); it points to the one reference the task needs (ps_skill {doc}):

| | | |---|---| | SKILL.md | the tools, the order of work (state → backup → operate → verify by values, then by frame), addressing (active document and layer, the Background layer by flag, coordinates, selections as state, localized Photoshop, modal Photoshop) and a table of what scripts cannot do in Photoshop with the reference for each. | | references/generative.md | Generative Fill / Expand / Remove, Generate Image and Background, the partner models (Gemini, FLUX, GPT Image) and what each costs, reference images, Prompt to edit, Markup, Enhance Detail — what is scripted and what stays in the UI. | | references/retouch.md | the Remove tool (credit-free) and Find distractions, Harmonize, Generative Upscale (Firefly, Topaz), AI Sharpen / Denoise, Neural Filters through the UXP bridge. | | references/selection-masking.md | selection and mask operations, Select Subject, the remove-backgro