Agent-S

{
  "agent_id": "agent_s",
  "name": "Agent-S",
  "url": "https://github.com/simular-ai/Agent-S",
  "logo": "https://agentery.com/logos/CP-AZC76V-256.png",
  "niche": "desktop-automation",
  "category": "developer-tools-infra",
  "short_summary": "autonomous GUI-based computer use: performs complex tasks on a computer by interacting with the screen like a human (clicking, typing), with optional local cod…",
  "task_performed": "autonomous GUI-based computer use: performs complex tasks on a computer by interacting with the screen like a human (clicking, typing), with optional local code execution",
  "inputs_accepted": [
    "natural language task instructions via CLI",
    "screenshots/GUI state",
    "optional local code environment inputs"
  ],
  "outputs_produced": [
    "GUI actions (mouse/keyboard control)",
    "executed Python/Bash code",
    "completed computer tasks"
  ],
  "integrations_available": [
    "OpenAI",
    "Anthropic",
    "Azure OpenAI",
    "Gemini",
    "Open Router",
    "vLLM",
    "Hugging Face Inference Endpoints",
    "UI-TARS"
  ],
  "protocols_or_interfaces": [],
  "industry_fit": [
    "general business",
    "developer tools",
    "automation"
  ],
  "autonomy_level": "agentic",
  "human_approval_needed": "unclear",
  "pricing_model": "free",
  "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": "2026-07-26T05:16:17.581Z",
    "amount": null,
    "display": null,
    "plans": [],
    "source": "render+llm"
  },
  "trust_or_rating_signal": [
    "11.9k GitHub stars",
    "1.4k forks",
    "ICLR 2025 Best Paper Award",
    "COLM 2025 paper acceptance",
    "SOTA on OSWorld (72.6%, surpassing human-level)",
    "published research papers (S1, S2, S3)"
  ],
  "evidence_quality": "high",
  "entity_type": "agent",
  "regulated_data_suitability": "unclear",
  "evidence_urls": [
    "https://github.com/simular-ai/Agent-S",
    "https://simular.ai"
  ],
  "last_checked": "2026-06-16",
  "how_to_connect": {
    "website": "https://github.com/simular-ai/Agent-S",
    "docs": "https://github.com/documentation",
    "mcp": {
      "endpoint": "https://github.com/mcp.json",
      "transport": "http",
      "config_snippet": "{\n  \"mcpServers\": {\n    \"agent_s\": {\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": 647,
    "uptime_7d": 1,
    "checked_at": "2026-07-28T02:30:37.802Z",
    "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."
}