{
  "schema_version": "1.1",
  "id": "s4:https://vercel.com/changelog/run-cursor-cloud-agents-vercel-sandbox",
  "slug": "run-cursor-cloud-agents-vercel-sandbox-0l3bo4n",
  "url": "https://feed7.dev/p/run-cursor-cloud-agents-vercel-sandbox-0l3bo4n",
  "title": "Cursor Cloud Agents can now run in Vercel Sandbox",
  "why_included": "Cursor Cloud Agents can execute inside per-request Vercel microVMs, giving teams control over isolation, credentials, retries, and worker lifecycle without operating persistent VMs.",
  "summary": "Cursor now lets Cloud Agents run on **Vercel Sandbox** through its Self-Hosted Machines APIs. Cursor retains the harness and inference loop, while each request gets an isolated **Firecracker microVM**; Vercel Functions and Workflow handle queues, provisioning, monitoring, retries, and cleanup.",
  "practical_implication": "This separates agent orchestration from execution infrastructure. Teams needing tighter runtime control can use scale-to-zero workers and short-lived, user-scoped credentials instead of maintaining a standing VM pool.",
  "agent_context": "Cursor now lets Cloud Agents run on **Vercel Sandbox** through its Self-Hosted Machines APIs. Cursor retains the harness and inference loop, while each request gets an isolated **Firecracker microVM**; Vercel Functions and Workflow handle queues, provisioning, monitoring, retries, and cleanup.\n\nThis separates agent orchestration from execution infrastructure. Teams needing tighter runtime control can use scale-to-zero workers and short-lived, user-scoped credentials instead of maintaining a standing VM pool.\n\nSelf-Hosted Machines requires a **Cursor Enterprise** plan, and the source describes a reference implementation rather than its operating cost or performance under load.",
  "source": {
    "name": "Vercel",
    "url": "https://vercel.com/changelog/run-cursor-cloud-agents-vercel-sandbox",
    "published_at": "2026-09-03T15:00:00.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Engineering Post",
  "layer": "infra",
  "domains": [
    "coding",
    "security"
  ],
  "topics": [
    "cloud-agents",
    "sandboxing",
    "coding-agents"
  ],
  "verification": {
    "status": "official_source",
    "label": "Official Source",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "Self-Hosted Machines requires a **Cursor Enterprise** plan, and the source describes a reference implementation rather than its operating cost or performance under load."
  ],
  "connected_context": {
    "meaning": "This turns the brain-versus-hands separation into a concrete Cursor deployment: Cursor keeps orchestration while Vercel supplies per-request isolated execution and lifecycle machinery. It reinforces per-agent sandboxing and user-scoped identity as practical controls, while narrowing applicability to Enterprise customers. The reference architecture still leaves load performance and operating economics unproven.",
    "corpus_size": 691,
    "generated_at": "2026-09-05T10:06:14.452Z",
    "connections": [
      {
        "title": "Scaling Managed Agents: Decoupling the brain from the hands",
        "source_name": "Anthropic",
        "source_url": "https://www.anthropic.com/engineering/managed-agents",
        "feed7_url": "https://feed7.dev/p/managed-agents-00zpz67",
        "reason": "Both separate the agent harness from replaceable execution environments, independently reinforcing decoupled orchestration and sandbox infrastructure as a managed-agent architecture."
      },
      {
        "title": "Give every agent in Herdr its own Vercel Sandbox",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/changelog/give-every-agent-in-herdr-its-own-vercel-sandbox",
        "feed7_url": "https://feed7.dev/p/give-every-agent-in-herdr-its-own-vercel-sandbox-1p4i553",
        "reason": "Both assign isolated Vercel sandboxes to individual agent work, but Herdr additionally returns reviewable Git patches, whereas the Cursor signal focuses on hosted provisioning and lifecycle operations."
      },
      {
        "title": "Tethered: Our Agents Are Us — Shu Fang, Two Sigma",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=wCIYViPd4SU",
        "feed7_url": "https://feed7.dev/p/tethered-our-agents-are-us-shu-fang-two-sigma-0c8u33i",
        "reason": "Cursor’s short-lived user-scoped credentials align with Two Sigma’s emphasis on preserving real-user attribution, though the candidate also adds trace and egress controls not established here."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-09-03T15:00:00.000Z",
  "modified_at": "2026-09-03T15:00:00.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/run-cursor-cloud-agents-vercel-sandbox-0l3bo4n",
    "json": "https://feed7.dev/p/run-cursor-cloud-agents-vercel-sandbox-0l3bo4n.json",
    "markdown": "https://feed7.dev/p/run-cursor-cloud-agents-vercel-sandbox-0l3bo4n.md"
  }
}