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.
git clone --depth 1 https://github.com/isac322/kwin-mcpWrote 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.
[](https://agentmods.dev/plugins/isac322/kwin-mcp/claude-code)<a href="https://agentmods.dev/plugins/isac322/kwin-mcp/claude-code"><img src="https://agentmods.dev/badge/plugins/isac322/kwin-mcp/claude-code/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.
<a href="https://agentmods.dev/plugins/isac322/kwin-mcp/claude-code"><img src="https://agentmods.dev/badge/plugins/isac322/kwin-mcp/claude-code.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
kwin-mcp scanned grade A with 0 findings 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 2d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
{
"$schema": "https://claude.ai/schemas/plugin.json",
"name": "kwin-mcp",
"version": "1.0.0",
"description": "MCP server + skill for Linux KDE Plasma 6 Wayland GUI automation. Provides 29 MCP tools (sessions, apps, screen capture, AT-SPI2 elements with refs, windows and virtual desktops, pointer, keyboard, touch, clipboard, D-Bus and Wayland escape hatches) plus the kwin-desktop-automation skill that guides session-mode selection, observation/action sequencing, and platform pitfalls.",
"author": {
"name": "Byeonghoon Yoo",
"email": "[email protected]",
"url": "https://github.com/isac322"
},
"homepage": "https://github.com/isac322/kwin-mcp",
"repository": "https://github.com/isac322/kwin-mcp",
"license": "MIT",
"keywords": [
"mcp",
"model-context-protocol",
"kwin",
"kde",
"plasma",
"wayland",
"gui-automation",
"desktop-automation",
"linux",
"linux-desktop",
"ai-agents",
"accessibility",
"at-spi2",
"libei",
"headless-testing",
"live-desktop",
"claude-code"
]
}
What it installs
The manifest is a name and a version. 1 skill travel with it, and installing the plugin installs all of them — 140 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- 2d ago Changed 5f38ebb4c0ab
- 3d ago Changed 3f48b71848c6
- 6d ago Changed f5ec6e6c3c4a
- 20d ago Changed 125d1ab43274
- 21d ago Changed 4bda2ea84b47
- 23d ago Changed f5e302cd4ac2
- 1mo ago Changed 8debfef64456
- 1mo ago First seen · 34 lines scan A 0bffb8b1a3ba
kwin-mcp is a plugin published in the GitHub repository isac322/kwin-mcp (70 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other plugins, from other repositories
screenhand
Desktop and browser automation for Claude Code. Control any macOS/Windows app, automate social media, run QA tests, edit video, design in Figma/Canva, scrape the web, and orchestrate multi-agent workflows — 111 MCP tools.
axstream
Fast, verified computer use for coding agents: read the screen as text (200ms), execute batched verified actions, and compile any task done once into a macro that replays in seconds with zero model calls.
qa-explore
A team of AI agents that test any web app like human QA: explore the live app, screenshot and visually judge rendering + data, capture trace/HAR/console/video evidence, adversarially verify findings, learn from rejected ones, codify them into a self-growing Playwright/Cypress suite — and close the loop by filing…
offstage-qa
Background QA for macOS apps via the offstage harness: AX perception, semantic port actuation, canonical pixel goldens, ground-truth verification.
sniff
Autonomous QA scanner via sniff-qa: walks your running app in a real browser and reports real bugs (broken pages/links, console/network errors, broken forms, state-loss, empty/placeholder data, bad loading/error states, responsive + a11y) with reproduction proof, severity, confidence, and a fix. No API key.
gnome-ui-mcp
GNOME Wayland desktop automation via AT-SPI discovery and Mutter input.