AI/ML●●●Banger
GhostDesk – MCP server giving AI agents a full virtual Linux desktop
Gives AI agents a full Linux desktop with human-like mouse movement to bypass bot detection.
WizardryZero to One
maltyxxx
503mo ago
Back to browse
GitHub Repository
MCP server for the Linux desktop 🐧🤖
2 starsRust
Desktopmcp – MCP server for the Linux desktop
by varbhat·Jun 24, 2026·2 points·1 comment
AI Analysis
●●●BangerWizardryBig BrainZero to One
AT-SPI semantic UI access beats pixel-scanning for desktop automation.
Strengths
- •144 tools across five categories show serious implementation depth.
- •AT-SPI accessibility tree lets AI find elements semantically, not via OCR.
- •XDG Portals ensure permission dialogs work across GNOME, KDE, and Sway.
Weaknesses
- •Linux-only limits audience compared to cross-platform alternatives.
- •MCP ecosystem still emerging, adoption uncertain.
Category
Target Audience
Developers building AI agents for desktop automation
Similar To
Computer Use API · Adept AI · OpenInterpreter
Similar Projects
Developer Tools●●Solid
Give Agents Isolated Linux Sandboxes via MCP [Kilntainers]
MCP sandbox isolation for agents; E2B/Modal/Docker/WASM backends already exist separately.
Niche GemBig BrainShip It
scosman
104mo ago
Developer Tools●●●Banger
OpenBrowser MCP: Give your AI agent a real efficient browser
3.2x–6x fewer tokens than Playwright/Chrome DevTools MCP via code-first architecture.
WizardrySolve My ProblemZero to One
billy-enrizky-1
204mo ago
AI/ML●●Solid
Hahooh – Give AI agents the power to build their own MCP tools
WordPress for MCP servers, but the agent-as-builder vision feels premature.
Ship ItBold Bet
hahooh
201mo ago
AI/ML●●Solid
Coding Tools MCP – Give any LLM agent the ability to code
Model-neutral MCP server with Docker sandbox and SWE-bench evals for any coding agent.
Big BrainNiche Gem
xytom
205d ago
Security●●Solid
TrustVector – Trust evaluations for AI models, agents, & MCP
The clever bit is the evidence-first, CVSS-style reweightable trust vector: each evaluation is a structured JSON with sources and confidence so you can compute custom scores or ingest them into tooling. The public methodology and 100+ entries make it immediately useful, though it needs more automation/CI integrations and live testing pipelines to feel like an operational assurance platform rather than a curated directory.
Niche GemSolve My Problem
hckdisc
214mo ago