Agent Sandbox Taxonomy
ActiveDescription
An open taxonomy and scoring framework for evaluating AI agent sandboxes with 7 defense layers and 7 threat models.
Related Projects
Daytona
72.4k · TypeScript
Daytona provides secure development-environment infrastructure for coding agents and automation workflows, serving as a runtime base for remote execution tasks.
dev-environmentsandboxcoding +1
MasterBlaster
71 · Go
Platform to create, manage, and orchestrate stereOS AI agent sandboxes with secure isolated execution environments.
sandboxagent-runtimeisolation +2
Jailoc
17 · Go
Jailed Docker environments with network isolation for AI agents to execute code safely in isolated containers.
sandboxdockernetwork-isolation +2
Spring AI Playground
179 · Java
Safe local execution layer for AI agent tools to build, validate, and publish MCP tools with a no-password secure runtime.
spring-aimcpagent-tools +2