Anthropic Sandbox Runtime vs CAPE Sandbox

Anthropic Sandbox Runtime (srt) is a free, open-source research preview that applies filesystem and network restrictions to processes at the operating-system level, without requiring a container. It can be used as a CLI tool or library to sandbox agents, local MCP servers, bash commands and other processes. Filesystem writes and network access are denied by default, with configuration options to allow paths and domains. The project documents macOS, Linux and Windows support, using sandbox-exec, bubblewrap or a dedicated local sandbox account with Windows Filtering Platform and filesystem ACLs, respectively. Windows is marked alpha. Install it from npm with `npm install -g @anthropic-ai/sandbox-runtime`. Linux requires bubblewrap, socat and ripgrep; macOS requires ripgrep, and Windows needs a one-time elevated setup step. On macOS, srt can provide real-time alerts from the system sandbox violation log. Linux bubblewrap has no built-in violation reporting, and Linux network filtering relies on proxy environment variables, which some programs may ignore. The preview is developed for Claude Code; its APIs and configuration formats may evolve.

CAPE Sandbox is a free, open-source malware sandbox that executes malicious files in isolation while recording behavior and forensic artifacts. It captures API-hook activity, file changes, network traffic as PCAP, desktop screenshots, and full memory dumps. Malware classification can use YARA scans of unpacked payloads, Suricata scans of network captures, and behavioral signatures from API-hook output. CAPE also performs dynamic unpacking and extracts malware configuration through static and dynamic analysis. Its programmable debugger supports custom unpacking and configuration extractors, anti-sandbox countermeasures, and instruction traces. Configuration parsing works with CAPE’s framework and the listed RATDecoders, DC3-MWCP, MalDuck, and MaCo projects. Documentation covers a REST API, distributed deployment, and endpoints for tasks, reports, files, PCAPs, and machines. Additional integrations include Box-js, Curtain, LibreNMS, and Suricata. The project recommends Ubuntu 24.04 LTS, Windows 10 or Windows 11 23H2 targets, and KVM. Active unpacking is off by default because it may affect detonation quality.