agentscope-ai/agentscope

Build and run agents you can see, understand and trust.

What it solves

AgentScope is a production-ready framework designed to build agentic applications by leveraging the reasoning and tool-use capabilities of LLMs rather than relying on strict prompts or rigid orchestrations. It provides the necessary abstractions to scale from a single agent to complex multi-agent teams and full-scale applications.

How it works

It operates through two primary layers:

  1. SDK Layer: Allows developers to compose agents using building blocks like ReAct loops, toolkits (including MCP servers and Python tools), and a unified event system for streaming multimodal content. It includes middleware for context compression and permission systems for Human-in-the-Loop (HITL) control.
  2. Agent Service: A FastAPI-based backend that transforms agents into multi-tenant applications. This layer handles session isolation, RAG services, persistence via SQL/NoSQL, and integration with IM platforms like Discord and Feishu (Lark).

Who it’s for

Developers building AI agents and multi-agent systems that require production-grade features like multi-tenancy, secure code execution in sandboxes, and integration with external communication channels.

Highlights

  • Comprehensive Tooling: Built-in support for coding tools (shell, file edit) and integration with MCP & Skill Hubs.
  • LMM Support: Compatible with major providers including OpenAI, Anthropic, Gemini, DeepSeek, and Ollama.
  • Flexible Execution: Supports isolated workspaces (Docker, E2B, K8s) and background task offloading to handle long-running tools.
  • Multi-Agent Orchestration: Features leader-worker patterns for task planning and coordination.
  • Real-time Capabilities: Support for full-duplex realtime APIs and voice agents.

Related

  • Project
    agentscope-ai/agentscope-java

    A production-ready Java framework for building distributed, enterprise-grade AI agents with built-in support for long-running tasks, secure sandboxing, and session recovery.

  • Project
    scaleapi/scale-agentex

    A framework for building and deploying AI agents that scales from simple chatbots to fully autonomous systems, supporting both synchronous and asynchronous workflows.

  • Project
    SciSharp/BotSharp

    An open-source .NET framework for building AI bot platforms that allows enterprise developers to integrate LLMs and multi-agent systems into business applications.

  • Project
    promptise-com/Foundry

    A full-stack agentic engineering framework that provides a unified system for building, reasoning, and deploying secure, multi-tenant AI agents with integrated tools and runtime.

  • Project
    MervinPraison/PraisonAI

    PraisonAI is a framework for building and deploying autonomous AI agents and multi-agent workflows, featuring a structured five-layer stack for prompt, context, harness, loop, and graph management.