OpenSandbox sandbox backend for LangChain Deep Agents - run agent-generated code and file operations in an isolated OpenSandbox environment.
-
Updated
Aug 17, 2026 - Python
OpenSandbox sandbox backend for LangChain Deep Agents - run agent-generated code and file operations in an isolated OpenSandbox environment.
An agent-native sandbox runtime: run many isolated sandboxes (beds) from one process, with an OpenSandbox-compatible HTTP API.
Self-hosted, multi-tenant AI agents platform built on deepagents, LangGraph & assistant-ui. Bring your own models (Anthropic, OpenAI, AWS Bedrock, Gemini), connect MCP tools, and give each user private memory, skills & an OpenSandbox-isolated workspace. FastAPI + React 19. Human-in-the-loop approvals built in.
MCP server for running heterogeneous AI agents and shell commands in local disposable OpenSandbox containers
Rewrite the core parts of OpenSandbox
OpenSandbox-backed sandbox provider for AI SDK Harness agents.
以Openclaw架构思想构建的基于Deep Agents、MCP、OpenSandbox、FastAPI 与 Vue 3 的多用户智能采购协作平台,支持多 Agent 编排、异步任务、SSE 流式交互、HITL 审批、Opensandbox沙箱隔离与 ERP 数据分析。
Sandboxed demo of crescendo attacks against LLM agents, with disk-fill, exfiltration, and deletion scenarios.
Opinionated Effect 4 template for secure, durable AI agent applications
OpenSandbox support for the Agno agent framework
OpenSandbox support for the Mastra agent framework
Transparent C scan-debate-proof POC using OpenSandbox, gVisor, and AddressSanitizer
To associate your repository with the opensandbox topic, visit your repo's landing page and select "manage topics."