Deskhand runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 7 Nov 2026.
Safe KDE Plasma (Wayland) desktop control for AI agents: OCR-checked input, approval panel.
We read the source, 6 h ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
if tur == "komut" and subprocess.run(deger, shell=True, capture_output=True, timeout=30).returncode == 0:
if tur == "komut" and subprocess.run(deger, shell=True, capture_output=True, timeout=30).returncode == 0:
with open(os.path.join(sd, f"{ad}.service"), "w") as f:
return __import__(f"deskhand.{ad}", fromlist=["_"])
env = dict(os.environ, DOTOOL_XKB_LAYOUT=duzen)
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
Safety-gated computer use for AI agents: screen OCR and vision, mouse, keyboard, window control.
Documents and web for AI agents: PDF parse, OCR, web extract, invoice checks. x402, no API key.
Windows screenshots, OCR, and redacted screen recording for AI agents. No desktop app needed.
Web extraction, OCR (Japanese-strong), invoice and company data for AI agents. x402, no API key.
Score documents and check agent work — signed PactCheck Receipts: proof-of-review for agents.
Evidence-backed UBL, PDF, and image invoice checks before approval or payment.
Screen observation, OCR and permission-controlled app input over MCP for Apple silicon Macs.
MCP server for browser and desktop automation: OCR, image recognition, real mouse and keyboard input
Answers built from our own checks of this server.