Releases: alibaba/open-agent-auth
Releases · alibaba/open-agent-auth
v0.1.0-beta.1
🎉 Release v0.1.0-beta.1
Open Agent Auth is now in public beta — feature-complete, actively evolving, and ready for you to explore.
🚀 Beta Philosophy
This release delivers core capabilities with APIs under active refinement. It demonstrates our core philosophy, protocol implementation, and architectural design for secure AI agent operations. We are actively improving based on real-world usage and community feedback as we progress toward a production-ready 0.1.0 release.
✨ Highlights
Core Features
- Agent Operation Authorization protocol implementation based on IETF Draft
- Three-layer cryptographic identity binding (User-Workload-Token)
- Dynamic policy registration with OPA integration
- WIMSE workload identity pattern for request-level isolation
- Five-layer security verification mechanism
- W3C VC-based semantic audit trail
Integration
- Spring Boot 3.x autoconfiguration
- MCP (Model Context Protocol) adapter
- OAuth 2.0 & OpenID Connect compliance
Quality
- Test coverage > 80%
- Comprehensive sample applications
- Complete documentation
📦 Quick Start
Run the sample project with Mock LLM:
git clone https://github.com/alibaba/open-agent-auth.git
cd open-agent-auth
mvn clean package -DskipTests
cd open-agent-auth-samples
./scripts/sample-start.sh --profile mock-llm
# Access the agent interface
open http://localhost:8081For QwenCode integration, see the Quick Start Guide.
🎯 What's Next
We are focused on stabilizing the core framework and delivering additional features for the 0.1.0 production release. Planned enhancements include:
- Stabilization and finalization of core external APIs and configurations
- Documentation improvements and bug fixes
- Additional security hardening
- Authorization Discovery for dynamic server negotiation
- Agent-to-Agent Authorization with delegation support
- OpenAPI Adapter for REST API integration
Making AI Agent Operations Safer, More Controllable, and More Traceable