{
  "schema_version": "1.0",
  "id": "archive:https://www.anthropic.com/engineering/managed-agents",
  "slug": "managed-agents-00zpz67",
  "url": "https://feed7.dev/p/managed-agents-00zpz67",
  "title": "Scaling Managed Agents: Decoupling the brain from the hands",
  "why_included": "Anthropic details Managed Agents, a hosted long-horizon agent service that separates the harness from its sandboxes — stateless brains, replaceable containers, and a 60% drop in p50 time-to-first-token.",
  "summary": "**The gist** Anthropic describes **Managed Agents**, a hosted service for long-horizon agent work that splits an agent into three virtualized parts: session log, harness, and sandbox. Harnesses run **stateless**, calling sandboxes through a generic **execute()** interface, so containers become disposable. Because the brain no longer waits on container provisioning, **p50 time-to-first-token dropped about 60%** and p95 over 90%.",
  "practical_implication": "**Why it matters** The pattern is reusable if you run your own agents: keep durable state in a session log outside the context window instead of irreversibly compacting it, keep credentials out of sandboxes (OAuth tokens stay in **external vaults behind MCP proxies**), and let **one brain drive multiple execution environments**. Harnesses can also deploy inside your own VPC.",
  "agent_context": "**The gist** Anthropic describes **Managed Agents**, a hosted service for long-horizon agent work that splits an agent into three virtualized parts: session log, harness, and sandbox. Harnesses run **stateless**, calling sandboxes through a generic **execute()** interface, so containers become disposable. Because the brain no longer waits on container provisioning, **p50 time-to-first-token dropped about 60%** and p95 over 90%.\n\n**Why it matters** The pattern is reusable if you run your own agents: keep durable state in a session log outside the context window instead of irreversibly compacting it, keep credentials out of sandboxes (OAuth tokens stay in **external vaults behind MCP proxies**), and let **one brain drive multiple execution environments**. Harnesses can also deploy inside your own VPC.\n\n**Watch out** This is architecture, not benchmark: no **cost comparisons**, scaling limits, or **session persistence guarantees** are given, and failure recovery is described in principle rather than measured.",
  "source": {
    "name": "Anthropic",
    "url": "https://www.anthropic.com/engineering/managed-agents",
    "published_at": null
  },
  "source_class": "blog_post",
  "content_type": "Engineering Post",
  "layer": "infra",
  "domains": [
    "coding"
  ],
  "topics": [
    "cloud-agents",
    "sandboxing",
    "harness-engineering"
  ],
  "verification": {
    "status": "official_source",
    "label": "Official Source",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "This is architecture, not benchmark: no **cost comparisons**, scaling limits, or **session persistence guarantees** are given, and failure recovery is described in principle rather than measured."
  ],
  "lifecycle": "Current",
  "published_at": null,
  "modified_at": null,
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/managed-agents-00zpz67",
    "json": "https://feed7.dev/p/managed-agents-00zpz67.json",
    "markdown": "https://feed7.dev/p/managed-agents-00zpz67.md"
  }
}