Loading post…
ModalAnnouncement
Introducing Claude Managed Agents with Modal Sandboxes
Summary
Anthropic launched Claude Managed Agents, a suite of composable APIs that separate the agent loop from code execution for building cloud-hosted agents at scale. Modal has collaborated with Anthropic to provide a first-class integration allowing all tool calls to run in self-hosted Modal Sandboxes, which offer fast cold-starts, custom images, snapshotting techniques, and cost-efficient burst pricing. These sandboxes also provide network security through short-lived connect tokens and support upwards of 100,000 concurrent sandboxes per customer. DoorDash is exploring Managed Agents as an accelerator for its merchant intelligence system, while Blend uses the integration to enable secure, scalable agent-assisted triage across complex banking environments. The integration gives teams isolation, cost efficiency, and control over observability, failure handling, and scale while maintaining the convenience of Anthropic’s managed agent loop.
Context
Anthropic’s Managed Agents run the agent loop externally, while Modal Sandboxes provide isolated, customizable compute for tool execution.
Approach / What changed
Modal Sandboxes integrate with Claude Managed Agents so that tool calls run in a self-hosted Modal Sandbox, while Anthropic manages the agent loop and session state via a REST API.
Takeaways
Managed Agents separate the agent loop from code execution, keeping the model loop outside the execution environment.
Modal Sandboxes provide custom images, fast cold-starts, volumes, directory snapshots, filesystem snapshots, and memory snapshots for persistence across runs.
Burst pricing on Modal Sandboxes means users pay only for what they use, with no minimums and no idle costs between runs.
FAQ
How do Modal Sandboxes secure tool access?
Modal Sandboxes use short-lived connect tokens that give clients authenticated, direct access to running sandbox services without exposing them to the public internet.
What are the differences between the Agent SDK and Managed Agents?
The Agent SDK allows building the agent loop inside your own process and infrastructure, while Managed Agents is a fully hosted REST API where Anthropic runs the agent loop and manages session state.
What scale does Modal Sandboxes support for Managed Agents?
Modal Sandboxes support upwards of 100,000 concurrent sandboxes per customer, with generous resource configurations including CPU, memory, and GPU access.
Source and freshness
- Source
- modal.com
- Published
- Added to Yomu
Summaries are generated from the original by a language model. Check the original before relying on it.
Keep reading
Related reading
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
The best engineering reads, once a week
One useful email. Unsubscribe anytime.