Skip to main content
Back to trending

agentscope

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

  • agent
  • chatbot
  • large-language-models
  • llm
  • llm-agent
  • mcp
  • multi-agent
  • multi-modal
  • react-agent
  • realtime-agent
  • voice-agent
  • voice-assistant
View on GitHub
Stars
33,139
Forks
3,639
+ today
+3
Created
2y

Ranking data as of October 12, 2026 (UTC).

Star History

Today, hour by hour

33,139 stars
03

Overview

Build and run agents you can see, understand and trust. It ranks #100 on GitTiger, gaining +3 stars on October 12, 2026 (UTC).

The project is written in Python and has 3,639 forks. It was created 2y ago.

AI-generated summary of the repository README — may contain errors.

What it is

`agentscope-ai/agentscope` describes a toolkit for building and running agents with an emphasis on making their behavior understandable and trustworthy. The repository is categorized under MCP. The supplied README preview references AgentScope 2.0 and includes some features marked experimental.

Capabilities

The preview says an experimental agent service supports real-time voice conversations over WebRTC, with text and voice sharing session history. It also mentions experimental support for running multi-step tasks as standard operating procedures. These details come from a truncated preview, so they should be treated as a partial view rather than a complete feature inventory.

Getting started

No installation or configuration instructions are included in the supplied evidence. Consult the repository README and documentation for current setup steps, version details, and the status of experimental features.

Installation
git clone https://github.com/agentscope-ai/agentscope.git
cd agentscope
# see README for setup

Frequently Asked Questions

What is AgentScope intended to help build?

The description presents it as a toolkit for building and running agents.

Does the supplied preview mention voice?

Yes. It describes an experimental service for real-time WebRTC voice conversations with shared text and voice session history.