A production-ready, multi-agent LLM application framework that provides fine-grained control over tools, multi-tenancy, and safe code execution environments. It requires coding and uses external LLM providers for inference.
Project overview
The framework supplies essential abstractions for complex, production-grade deployments, including event-driven human-in-the-loop support and multi-tenancy serving.
Project type
AI Agent · Infrastructure
Use cases
Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers and AI engineers building multi-agent applications who need configurable permissions and multi-tenancy controls.
Teams requiring isolated sandbox workspaces for agent-driven code execution.
Key capabilities
Fine-grained, configurable control over tools and resources.
Runs tools and code in isolated environments, with built-in backends for local, Docker, E2B, OpenSandbox, and Daytona.
Limitations and risks
Requires an external LLM provider, such as DashScope qwen3.6-plus, which may incur paid service costs.
GPU requirements, minimum hardware, telemetry practices, and end-to-end data boundaries are not currently documented.
Getting started
Setting up the framework has a medium difficulty rating.
Prepare external model API credentials, such as DASHSCOPE_API_KEY, to enable inference.
Evidence and sources
GitHub project description: Build and run agents you can see, understand and trust.
README: AgentScope 2.0 is a production-ready, easy-to-use agent framework with essential abstractions that work with rising model capability and built-in support for .