screenmcp

screenmcp is a skill for Claude Code from shimondoodkin/screenmcp-claude-plugin. It costs 61 tokens per session (972 once invoked), scanned B, original, MIT.

A desktop-control skill for seeing and operating the computer’s graphical desktop. It can take screenshots, interact with windows, type text, use keyboard shortcuts, and work with the clipboard.

In plain words
What is it for?
Inspecting the screen, clicking or dragging, scrolling, entering text, pressing keys, managing windows, and reading or setting the clipboard.
Why use it?
It lets an AI agent perform tasks that require the actual desktop interface rather than only reading or editing project files.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions Claude Code.

Part of the screenmcp-claude-plugin plugin — 1 skill, 1 MCP server shipped together

Good fit Inspecting the screen, clicking or dragging, scrolling, entering text, pressing keys, managing windows, and reading or setting the clipboard.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/shimondoodkin/screenmcp-claude-plugin/screenmcp
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add shimondoodkin/screenmcp-claude-plugin --skill screenmcp
Clone the repo
git clone --depth 1 https://github.com/shimondoodkin/screenmcp-claude-plugin

Made for: Claude Code.

Or install screenmcp-claude-plugin, the plugin that ships this one along with the rest of its 1 skill, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for screenmcp

README.md
[![agentmods](https://agentmods.dev/badge/skills/shimondoodkin/screenmcp-claude-plugin/screenmcp/github.svg)](https://agentmods.dev/skills/shimondoodkin/screenmcp-claude-plugin/screenmcp)
Your own site
<a href="https://agentmods.dev/skills/shimondoodkin/screenmcp-claude-plugin/screenmcp"><img src="https://agentmods.dev/badge/skills/shimondoodkin/screenmcp-claude-plugin/screenmcp/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for screenmcp

Your own site · 80×15
<a href="https://agentmods.dev/skills/shimondoodkin/screenmcp-claude-plugin/screenmcp"><img src="https://agentmods.dev/badge/skills/shimondoodkin/screenmcp-claude-plugin/screenmcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 972 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00061 $0.00972
Opus 5 $0.00030 $0.00486
Sonnet 5 $0.00012 $0.00194
Haiku 4.5 $0.00006 $0.00097

Measured 9d ago against content hash 8c4fadf36e61, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade B, and why

screenmcp scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

`sudo apt install wmctrl xdotool`. Wayland is best-effort.
skills/screenmcp/SKILL.md · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ScreenMCP — Local Desktop Control

This plugin bundles a standalone stdio MCP server (screenmcp-local) that controls the computer Claude Code is running on — Windows, macOS, or Linux. No worker, relay, account, or network. Use it whenever the task means looking at or operating the real desktop GUI.

Setup (automatic)

The server's Python dependencies install themselves on first launch via bootstrap.py (mss, pynput, Pillow, pyperclip, and pygetwindow on Windows). You don't normally do anything.

If the mcp__screenmcp-local__* tools aren't available, the server failed to start — usually no Python on PATH, or a sandboxed install. Check its stderr, or install the deps manually:

pip install mss pynput Pillow pyperclip pygetwindow

Optional extras (only if the user needs them):

pip install opencv-python   # camera / list_cameras
pip install simpleaudio     # play_audio

Core loop: look, then act

  1. screenshot to see the screen before acting. Re-screenshot after actions that change the UI.
  2. Act with click, double_click, right_click, drag, scroll, type, press_key, hotkey (e.g. [ctrl, c]).
  3. Verify by screenshotting again.

When unsure where a click will land, use screenshot with the dots argument to paint estimated click positions and confirm before committing.

Coordinate space (important)

Screenshots default to 1456×819. All click/drag/scroll/move coordinates are in that same space and auto-scale to the real screen — so read positions off the screenshot you just took and pass them straight through. Override per call with max_width/max_height (0 = native pixels).

Model-aware sizing

Coordinate-bearing tools accept an optional model arg: claude | gemini | chatgpt. With no explicit max_width/max_height, it sizes the screenshot and the click space to that model's vision limits so the image and the clicks share one space. Since you are Claude, pass model: "claude" for best alignment. Explicit max_width/max_height always wins.

Read the full file on GitHub · 87 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 87 lines · 61 tokens per session scan B 8c4fadf36e61

Subscribe to this mod's changes

screenmcp is a skill published in the GitHub repository shimondoodkin/screenmcp-claude-plugin (0 stars, last pushed 2mo ago), licensed MIT. It adds 61 tokens to every session and 972 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens