agentscope-ai/agentscope-java

Build distributed, production-grade, long-running agents.

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 15 minutes ago
Added to GitGenius on September 12th, 2026
Created on September 23rd, 2025
Open Issues & Pull Requests: 875 (+0)
GitHub issues: Enabled
Number of forks: 1,347
Total Stargazers: 5,578 (+0)
Total Subscribers: 38 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 18.0 hours
Mean response time: 9.4 days
90th percentile: 22.4 days
Tracked items: 1,016

How this project is maintained

Around half of the issues opened in the past year never receive a reply. 95% of open issues come from outside the core team, so the backlog reflects real-world use rather than internal planning. 38% of tracked open issues have had no activity in three months. Only 6% of issues opened in the past year have been closed. Three people close 60% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 619
New in 7 days: 46
Closed in 7 days: 40
Avg open age: 94 days
Stale 30+ days: 448
Stale 90+ days: 223

Recent activity

Opened in 7 days: 44
Closed in 7 days: 34
Comments in 7 days: 10
Events in 7 days: 26

Top labels

  • bug (642)
  • enhancement (369)
  • area/core/agent (175)
  • area/harness (148)
  • area/core/tool (116)
  • area/ext/integration (101)
  • area/core/model (99)
  • area/core/memory (64)

Detailed Description

AgentScope Java is a framework for building distributed, production-grade, long-running agents in Java.

The framework addresses the challenge of deploying agents safely and reliably in enterprise environments. It provides a unified event system with typed events for real-time monitoring and human oversight, a permission system for controlling tool execution through approval workflows, and middleware hooks that allow flexible extension of the agent reasoning-acting loop. Tool execution can be isolated in sandboxed environments ranging from local execution to Docker, Kubernetes, or cloud-based sandboxes. For multi-agent scenarios, the framework supports orchestration patterns including agent spawning and message passing with real-time event forwarding between agents.

Teams building production agents should consider this framework if they need distributed deployment across multiple replicas with session recovery, persistent memory management across Redis, MySQL, PostgreSQL, and object storage systems, or human-in-the-loop control over agent actions. The tool suits long-running agent applications where observability, auditability, and safe tool execution are requirements. It requires Java 17 or later and is distributed through Maven Central.

The project maintains active development with regular updates to its core abstractions and production deployment patterns. Documentation is comprehensive and available in both English and Chinese. The framework includes built-in support for agent observability and evaluation workflows, suggesting ongoing investment in the operational lifecycle of deployed agents beyond initial development.