photoshop-mcp

MCP for Adobe Photoshop. 118 tools. Control from Cursor, Claude or custom LLMs.

Runs with mockssource: GitHubhomepageTypeScriptMITcommit a0b73188ba77

TypeScript, MIT licensed. The project labels itself: adobe, ai, ai assistant, claude, creative automation, cursor, extendscript and firefly.

photoshop-mcp runs, with stand-ins for the services it depends on. An Argusic agent installed it in 35 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

Project installs and builds cleanly; the MCP server starts and answers real JSON-RPC with 128 tools and 24 prompts, the standalone UI serves on 127.0.0.1 with 200 responses on / and token-protected /api/health, the 35-suite unit suite passes 254/254 (+1 skip) under the container's Node 18, and the only unimplemented path is live Adobe Photoshop itself, which is unavailable in a Linux container.

What the agent ended up with on a clean machine, in its own words. How this is measured

Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 92 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs with mocks
Argusic Score
92 / 100
cost of the verifying run
$0.10 (measured)
recorded runs
1
last tested
stars
660
forks
73
open issues
10
watchers
1
size
20 MB
created
last push

Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.

What was measured

From the run that decided this verdict. Every line is a recorded value, not an estimate.

Time lost to each failure while testing photoshop-mcp, drawn from the recorded run
Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

  • better-sqlite3 v12 prebuilt binary linked to libnode.so.109 ABI 109 segfaults under Node 20 (ABI 115), crashing 5 suite files15 minutes
  • npm install failed: @tailwindcss/oxide native binding missing (npm #4828 optional-deps bug)8 minutes
  • npm runtime mismatch: CLI/test scripts spawn 'npx' which resolves Node 18 under this PATH, but Node 20 crashes sqlite paths8 minutes
  • vitest 4 cannot start on Node 18.19.1 (node:util.styleText missing)3 minutes
Install time
35 minutes
Cold machine to finish
34 minutes
Errors hit and fixed
4 hit, 4 fixed with no human help
How the result was proved
npm run build exit 0; node dist/index.js answered initialize (photoshop-mcp 1.7.37), tools/list->128, prompts/list->24, photoshop_ping structured not-found over real stdio; curl GET / -> 200 and /api/health with token -> 200 on dist/ui/cli.js; vitest 254 passed, 1 skipped, 35/40 files; test-cursor-mcp-smoke and test-mcp-local exit 0; verify:pack ok (169 packed files); lint exit 0
Model tokens used
2,696,097
Exact commit tested
a0b73188ba77
Container image
sha256:cdd920bce783

What was not measured

The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.

Peak GPU memory
this ran on a CPU machine, so there was no GPU to sample
Behaviour over a long run
each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
Behaviour against real services
this run was verified against stand-in services only

The end of the recording

The last lines of the terminal session that decided this verdict, exactly as they were recorded.

**Errors (each diagnosed and fixed or confirmed unfixable)**
1. `@tailwindcss/oxide` binding missing → fix: install `[email protected]` (and `@rolldown/…
2. vitest 4 cannot start on the container's Node 18.19.1 (`node:util.styleText` needs ≥20) → workar…
3. `better-sqlite3` v12's bundled `build/Release/better_sqlite3.node` is linked to `libnode.so.109`…
4. Not-a-bug findings: 40-test full `vitest run` exits `0` on my box with 5 transient worker "error…
**Verified final state (only what I observed)**
- Full build green (`npm run build` exit 0); server binary launches (`node dist/index.js` prints "R…
- 254 unit tests pass + 1 skip across 35 files; the 5 sqlite-dependent files pass individually unde…
- Standalone UI: `node dist/ui/cli.js --port 5178 --no-open` → root `GET /` = 200 (HTML), `GET /api…
- `photoshop_ping` returns the structured not-found envelope; `photoshop_get_capabilities` returns …
**Bottom line for the ARGUSIC packet:** `install_succeeded: true`, `launch_succeeded: true`, `insta…
`ARGUSIC_RESULT: {"install_succeeded":true,"launch_succeeded":true,"install_minutes":35,"errors":[{…

Replay the whole session, every command from a clean machine to this point.

How it was tested

One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.

Strengths and limits

Measured facts, not opinions. How this is written.

What went well

  • Installed in 35 minutes.
  • Recovered from all 4 errors without a human stepping in, which says the failures are documented well enough to solve.
  • Licensed MIT, as reported by its host.

What did not, or is not known

  • Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
  • Measured once so far. A single attempt shows it can run, not that it runs reliably.

Also tested, in the same area

Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.

Run history

tested byattemptoutcomescorecost USDwhen
Argusic Runner1/3Runs with mocks92.000.10

Topics (from GitHub)

adobeaiai-assistantclaudecreative-automationcursorextendscriptfireflygenerative-aigraphic-designimage-editingmacosmcpmodel-context-protocolnpmphotoshopphotoshop-scriptingtypescriptvuewindows

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/photoshop-mcp.svg)](https://argusic.com/subject/photoshop-mcp)

Questions

Does photoshop-mcp run?
photoshop-mcp runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 35 minutes, hitting 4 errors on the way, and recorded the session.
How did Argusic test photoshop-mcp?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit a0b73188ba77. One attempt is recorded, and the full method is on the methodology page.
What did it cost to test photoshop-mcp?
The run that produced this verdict cost $0.10: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does photoshop-mcp take to install?
35 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
Does photoshop-mcp need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing photoshop-mcp?
4 things broke in the recorded run, and 4 were fixed without human help. Each one, and the time it cost, is listed on this page.
Where is the evidence for photoshop-mcp?
The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.

Discussion