AgentScope

What is AgentScope?

AgentScope 2.0 is an agent framework for teams that want to move from a prototype agent to a hosted service. It provides building blocks for messages, models, tools, planning, middleware, retrieval and long-term memory, and an Agent Service layer built on FastAPI that serves agents over HTTP to many users and sessions. Tool calls can be reviewed, paused for human approval and run in isolated workspaces such as the local filesystem, Docker, E2B or Kubernetes. A leader agent can spawn worker agents, run scheduled jobs and hand long tasks to the background, and channels can put agents on Feishu and Discord. Memory can connect to Mem0, ReMe and other vector databases. The documentation states no pricing, and the pricing URL returns a not-found page; the example backend needs Redis and the code is cloned from GitHub.

Use cases

• Build multi-agent applications
• Host agents as a multi-user HTTP service
• Run agent tools in sandboxes
• Add long-term memory to agents
• Schedule recurring agent jobs
• Connect agents to Feishu or Discord

Who it's for

• AI engineers
• Backend developers
• Agent framework users

Why use AgentScope

• Moving from a notebook agent to a hosted service needs little rewriting, since the Agent Service wraps the same agent code.
• Layered controls, from tool review to human approval to sandboxes, reduce the risk of agents touching your system.
• Per-user, per-agent and per-session isolation lets one deployment serve many tenants.
• Middleware hooks let you change agent behavior without editing the framework.

AgentScope features

• Agent, model, tool, context and plan building blocks
• Tool and agent middleware for changing runtime behavior
• Human-in-the-loop permission checks before tool calls
• Sandboxed tool execution on Docker, E2B, Kubernetes or local files
• FastAPI Agent Service with multi-tenant, multi-session management
• Agent teams with a leader and worker agents
• Cron scheduling and background task offloading
• Built-in RAG knowledge base service
• MCP and skill support, plus A2A protocol
• Session replay over server-sent events
• Experimental realtime voice over WebRTC

AgentScope integrations

• Mem0
• ReMe
• Docker
• E2B
• Redis
• Feishu
• Discord

How to get started with AgentScope

1. Clone the AgentScope repository from GitHub and start a local Redis.
2. Run the example Agent Service backend from examples/agent_service with Python.
3. Start the example React web UI and connect it to the backend.
4. Define your own agents, tools, models and workspace backend, then expose them with create_app.
5. Add your own authentication, then deploy the service and connect channels or schedules.

AgentScope limitations

• Version 2.0 is a breaking change from 1.0, so older code needs migration
• No built-in user authentication; you supply your own through a placeholder header
• Distributed deployment is marked work in progress and relies on Redis
• Several parts, such as pipelines and realtime voice, are labeled experimental

AgentScope pricing

AgentScope is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

AgentScope alternatives

The main AgentScope alternatives listed on Debuly are LangGraph, CrewAI, AutoGen, LlamaIndex, OpenAI Agents SDK and Agno. The table compares pricing and platforms; order is A–Z.

1. 

   Agno

   Python framework and AgentOS runtime for building agents, teams and workflows, then serving them as an API with a control plane.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit Agno (official site)
2. 

   AutoGen

   Microsoft's MIT-licensed Python and .NET framework for multi-agent AI apps, now archived; new projects are pointed to Microsoft Agent Framework.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit AutoGen (official site)
3. 

   CrewAI

   Multi-agent framework and platform for building AI agent crews in Python or a visual editor, then deploying and monitoring them.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit CrewAI (official site)
4. 

   LangGraph

   Low-level open-source orchestration framework and runtime for building stateful, long-running AI agents as graphs.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit LangGraph (official site)
5. 

   LlamaIndex

   Open-source framework plus the LlamaParse platform for parsing, extracting and indexing documents so LLM apps can use them.

   Developer platformsResearch and searchAutomation and agents
   FreemiumVisit LlamaIndex (official site)
6. 

   OpenAI Agents SDK

   Open-source Python framework for building multi-agent workflows with tools, handoffs, guardrails, tracing, sandboxes and voice agents.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit OpenAI Agents SDK (official site)

AgentScope FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from docs.agentscope.io · fetched 23 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.