# Sandbox Runtime (srt) — AI Coding Tool Review & Pricing

> Sandbox Runtime (srt) is Anthropic's lightweight OS-level sandbox, developed for Claude Code, that restricts the filesystem and network access of agents, local MCP servers, bash commands and arbitrary processes without a container. It is a research preview usable as a CLI or a library.

Source: https://ai.dosa.dev/tools/sandbox-runtime-srt · Updated Oct 28, 2026 · Category: Developer Productivity & Workflow · Published by dosa.dev

## At a glance

- **Tool**: Sandbox Runtime (srt)
- **Company**: Anthropic
- **Category**: [Developer Productivity & Workflow](https://ai.dosa.dev/category/productivity)
- **Pricing model**: Open Source
- **Official website**: https://github.com/anthropics/sandbox-runtime

## Key features

- Wrap any command with srt to enforce filesystem, network and Unix-socket restrictions
- Deny-by-default network allowlists enforced through built-in HTTP and SOCKS5 proxies
- Read and write path rules configured in ~/.srt-settings.json
- Native primitives: sandbox-exec on macOS, bubblewrap on Linux, a dedicated user plus Windows Filtering Platform rules on Windows
- Drop-in sandboxing for MCP servers in .mcp.json and real-time violation monitoring on macOS

## Pricing

**Model**: Open Source

Free and open source under the Apache-2.0 license, distributed on npm as @anthropic-ai/sandbox-runtime.

## Who it is for

- **Best for**: Developers and platform teams who want to run coding agents, MCP servers or untrusted scripts with tight filesystem and network limits on their own machines.
- **Not ideal for**: Teams needing hardened multi-tenant isolation or production guarantees, since it is a beta research preview rather than a VM or container sandbox.

## Verdict

A simple, container-free way to put guardrails around agents and MCP servers, backed by the same approach Anthropic uses in Claude Code.

## Recent updates

Actively released, with v0.0.81 published on October 20, 2026, and trending on GitHub in late October 2026.

## Frequently asked questions

### Is Sandbox Runtime (srt) free?

Yes - Sandbox Runtime (srt) is Open Source. Free and open source under the Apache-2.0 license, distributed on npm as @anthropic-ai/sandbox-runtime.

### Is Sandbox Runtime (srt) open source?

Yes - Sandbox Runtime (srt) is open source. Free and open source under the Apache-2.0 license, distributed on npm as @anthropic-ai/sandbox-runtime.

### Who is Sandbox Runtime (srt) best for?

Developers and platform teams who want to run coding agents, MCP servers or untrusted scripts with tight filesystem and network limits on their own machines.

### Who is Sandbox Runtime (srt) not ideal for?

Teams needing hardened multi-tenant isolation or production guarantees, since it is a beta research preview rather than a VM or container sandbox.

### What are the best Sandbox Runtime (srt) alternatives?

The closest Sandbox Runtime (srt) alternatives on ai.dosa.dev are Pieces for Developers, agenttrace, Vibe Kanban - all listed under Developer Productivity & Workflow.

### Who makes Sandbox Runtime (srt)?

Sandbox Runtime (srt) is developed by Anthropic. It is listed in the Developer Productivity & Workflow category on ai.dosa.dev.

## Top alternatives

- [Pieces for Developers](https://ai.dosa.dev/tools/pieces-for-developers) (Freemium) — Developers and teams who need to manage disparate workflow context, revisit past coding decisions, and maintain consistent documentation across multiple tools and projects.
- [agenttrace](https://ai.dosa.dev/tools/agenttrace) (Open Source) — Developers who need a privacy-conscious way to audit, monitor, and diagnose the behavior, costs, and performance of various AI coding agents locally.
- [Vibe Kanban](https://ai.dosa.dev/tools/vibe-kanban) (Free) — Solo developers and individual power users who need to run multiple AI coding agents in parallel while managing separate git worktrees and tasks locally.
- [ai-memory](https://ai.dosa.dev/tools/ai-memory) (Open Source) — Developers who frequently switch between multiple AI coding agents or machines and want a unified, persistent knowledge base that survives session ends and tool switches.
- [smolvm](https://ai.dosa.dev/tools/smolvm) (Open Source) — Running untrusted code, powering coding agents with persistent/resettable workspaces, and deploying portable VM-based workloads that require a real Linux kernel.
- [coop](https://ai.dosa.dev/tools/coop) (Open Source) — Security-conscious developers who want to run autonomous coding agents on real projects without risking host machine compromise.
- Full alternatives guide: https://ai.dosa.dev/tools/sandbox-runtime-srt/alternatives