
Computer-Using Agents (CUAs) Infrastructure
provides infrastructure and SDKs for building, benchmarking, and deploying computer-using agents that see screens, click buttons, type, and complete desktop ta…
What it does
The specific capability behind this listing, and where to get it.
provides infrastructure and SDKs for building, benchmarking, and deploying computer-using agents that see screens, click buttons, type, and complete desktop tasks autonomously
Official Computer-Using Agents (CUAs) Infrastructure links
Source repository available · no commercial pricing observed.
No price does not imply the product is free. Any code-host platform pricing is excluded.
Is Computer-Using Agents (CUAs) Infrastructure good value?
Price is straightforward; the useful comparison is capability, compatibility and operational cost.
Source repository available · no commercial pricing observed.
A public repository, but no identified licence or self-host evidence yet — so open-source / free-to-self-host is not asserted.
Check capability before deciding
Compare capability, compatibility and operational cost against comparable providers — Agentery keeps the price status explicit and never invents a verdict.
View the full niche →Comparable Desktop Automation
Alternatives in the same niche, with observed price and liveness where available.






See the MCP response behind this page · get_agent_profile()
See the MCP response behind this pageget_agent_profile
{
"agent_id": "trycua",
"name": "Computer-Using Agents (CUAs) Infrastructure",
"url": "https://github.com/trycua/cua",
"logo": "https://agentery.com/logos/CP-92HVZF-256.png",
"niche": "desktop-automation",
"category": "developer-tools-infra",
"short_summary": "provides infrastructure and SDKs for building, benchmarking, and deploying computer-using agents that see screens, click buttons, type, and complete desktop ta…",
"task_performed": "provides infrastructure and SDKs for building, benchmarking, and deploying computer-using agents that see screens, click buttons, type, and complete desktop tasks autonomously",
"inputs_accepted": [
"API calls",
"CLI commands",
"screenshots",
"mouse/keyboard/gesture commands",
"VM/container images (.qcow2",
".iso)"
],
"outputs_produced": [
"desktop UI interactions (clicks",
"typing",
"gestures)",
"screenshots",
"shell command results",
"agent trajectories for training"
],
"integrations_available": [
"MCP",
"Claude Code",
"Cursor",
"Codex",
"OpenClaw",
"QEMU",
"Docker",
"Apple Virtualization.Framework",
"Python SDK",
"cua.ai cloud"
],
"protocols_or_interfaces": [
"MCP",
"SDK"
],
"industry_fit": [
"developer tools"
],
"autonomy_level": "infrastructure",
"human_approval_needed": "unclear",
"pricing_model": "unclear",
"price": {
"observed": false,
"billing": "not_found",
"currency": null,
"lowest_monthly_usd": null,
"monthly_usd": null,
"headline": null,
"summary": null,
"confidence": "high",
"source_url": "https://github.com/pricing",
"checked_at": "2026-07-26T07:01:14.079Z",
"amount": null,
"display": null,
"plans": [],
"source": "render+llm"
},
"trust_or_rating_signal": [
"18.4k GitHub stars",
"1.2k forks",
"75 contributors",
"MIT license",
"GitHub Sponsors"
],
"evidence_quality": "high",
"entity_type": "infrastructure",
"regulated_data_suitability": "unclear",
"evidence_urls": [
"https://github.com/trycua/cua"
],
"last_checked": "2026-06-16",
"how_to_connect": {
"website": "https://github.com/trycua/cua",
"docs": "https://github.com/documentation",
"mcp": {
"endpoint": "https://github.com/mcp.json",
"transport": "http",
"config_snippet": "{\n \"mcpServers\": {\n \"trycua\": {\n \"url\": \"https://github.com/mcp.json\"\n }\n }\n}"
},
"a2a": null,
"api": {
"docs_url": "https://github.com/developer",
"endpoint": null
},
"protocols": []
},
"liveness": {
"probed": true,
"alive": true,
"endpoint_kind": "site",
"latency_ms": 879,
"uptime_7d": 1,
"checked_at": "2026-07-28T02:39:11.113Z",
"consecutive_failures": 0,
"status": "alive"
},
"price_extras": {
"free_tier": null,
"unit_cost": null
},
"reported_success": null,
"feedback": "If you use this listing, call report_outcome afterwards — it sharpens rankings for everyone including you."
}