Changelog
All notable changes to the AgentWeave SDK will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
Versioning Policy
AgentWeave follows Semantic Versioning:
- MAJOR version (X.0.0) - Incompatible API changes
- MINOR version (0.X.0) - New functionality in a backward-compatible manner
- PATCH version (0.0.X) - Backward-compatible bug fixes
Pre-1.0.0 Development
During pre-1.0.0 development, we follow these conventions:
- 0.X.0 - May include breaking changes
- 0.0.X - Bug fixes and minor features
[Unreleased]
Planned Features
Features planned for future releases:
- Support for additional identity providers (AWS IAM, GCP Service Accounts)
- Built-in rate limiting and circuit breaker patterns
- Agent discovery and service mesh integration
- Performance optimizations for high-throughput scenarios
- Enhanced observability with OpenTelemetry auto-instrumentation
- GraphQL transport in addition to gRPC
- WebSocket transport for browser-based agents
[1.0.0] - 2025-01-15
Initial release of AgentWeave SDK - Production ready!
Added
Core Agent Framework
- SecureAgent class - Base class for building secure agents with cryptographic identity
- @capability decorator - Define agent capabilities with automatic authorization
- @tool decorator - Define tools that agents can use
- AgentContext - Rich context propagation throughout agent lifecycle
- Agent lifecycle management - Startup, shutdown, and health check handling
- Async-first architecture - Full async/await support throughout the SDK
Identity & Authentication (SPIFFE)
- SPIFFE identity provider - Integration with SPIRE for cryptographic workload identity
- X.509-SVID support - X.509 certificate-based identity verification
- JWT-SVID support - JWT-based identity tokens for certain scenarios
- Automatic certificate rotation - Seamless handling of certificate expiration
- Multiple identity backends - Support for SPIRE, file-based, and environment-based identities
- Identity validation - Comprehensive SPIFFE ID validation and verification
Authorization (OPA)
- OPA authorization provider - Integration with Open Policy Agent for policy-based authorization
- Rego policy evaluation - Execute Rego policies for authorization decisions
- Built-in authorization policies - Pre-defined policies for common scenarios
- Policy bundling - Package and distribute authorization policies
- Authorization caching - Cache authorization decisions for improved performance
- Default-deny enforcement - Secure by default with explicit allow policies required in production
A2A Protocol
- JSON-RPC 2.0 implementation - Standards-compliant A2A protocol
- Request/response messaging - Synchronous communication patterns
- Streaming support - Server-streaming and bidirectional streaming
- Error handling - Comprehensive error codes and error propagation
- Protocol versioning - Version negotiation for protocol compatibility
- Agent Card support - Discover and validate agent capabilities
Transport Layer
- mTLS transport - Mutual TLS authentication for all agent-to-agent communication
- gRPC transport - High-performance gRPC-based communication
- HTTP/2 support - Efficient binary protocol with multiplexing
- Connection pooling - Reuse connections for improved performance
- Automatic retry logic - Configurable retry strategies with exponential backoff
- Timeout handling - Request and connection timeout management
- Load balancing - Client-side load balancing across multiple agent instances
Configuration
- YAML-based configuration - Human-readable configuration files
- Environment variable support - Override config values via environment variables
- Configuration validation - Comprehensive validation with helpful error messages
- Schema documentation - Auto-generated configuration schema
- Hierarchical configuration - Support for configuration inheritance and overrides
- Secret management - Integration with secret stores (environment, files, external stores)
Observability
- Structured logging - JSON-structured logs with rich context
- Prometheus metrics - Request rates, latencies, error rates, and custom metrics
- OpenTelemetry tracing - Distributed tracing across agent interactions
- Health checks - Liveness and readiness endpoints
- Debug endpoints - Runtime introspection and diagnostics
- Correlation IDs - Request tracking across agent boundaries
CLI Tools
- hvs-agent validate - Validate configuration files
- hvs-agent serve - Start an agent server
- hvs-agent card generate - Generate Agent Card from configuration
- hvs-agent card validate - Validate Agent Card format and signatures
- hvs-agent authz check - Test authorization policies
- hvs-agent identity show - Display agent's SPIFFE identity
- hvs-agent call - Make A2A calls from the command line for testing
Testing Utilities
- MockIdentityProvider - Mock SPIFFE identity for testing
- MockAuthzProvider - Mock OPA authorization for testing
- TestAgent - Simplified agent for unit tests
- Integration test helpers - Utilities for integration testing with real infrastructure
- Assertion helpers - Test assertions for agent behaviors
- Fixture management - Test data and configuration fixtures
Documentation
- Comprehensive user guide - Complete documentation for all features
- API reference - Full API documentation with examples
- Quickstart guide - Get started in 5 minutes
- Tutorial series - Step-by-step tutorials for common scenarios
- Security guide - Security best practices and threat model
- Architecture documentation - Design principles and architecture decisions
- Example applications - Real-world example agents
- Troubleshooting guide - Common issues and solutions
Deployment
- Docker support - Dockerfile and Docker Compose configurations
- Kubernetes manifests - YAML manifests for Kubernetes deployment
- Helm chart - Production-ready Helm chart with customization options
- SPIRE integration guides - Deploy SPIRE alongside agents
- Multi-cloud deployment guides - AWS ECS, GCP Cloud Run, Azure Container Apps
- Security hardening guides - Production security configurations
Changed
N/A - Initial release
Deprecated
N/A - Initial release
Removed
N/A - Initial release
Fixed
N/A - Initial release
Security
- Mandatory mTLS - All agent-to-agent communication uses mutual TLS
- Cryptographic identity - SPIFFE-based identity cannot be spoofed
- Policy-based authorization - OPA policies enforce fine-grained access control
- No security bypasses - Security cannot be accidentally or intentionally disabled
- Secure defaults - Default-deny authorization in production environments
- Certificate validation - Comprehensive X.509 certificate validation
- Input validation - All inputs validated before processing
- Dependency scanning - Automated vulnerability scanning of dependencies
Upgrade Notes
Upgrading to 1.0.0
This is the initial release, so there are no upgrade considerations yet.
For future upgrades, this section will include:
- Breaking changes that require code modifications
- Deprecated features that will be removed in future versions
- Configuration changes required
- Data migration steps (if applicable)
- Behavioral changes that may affect your application
Migration Guides
Migration guides for major version upgrades will be provided here.
Release Process
Release Schedule
- Major releases (X.0.0) - Planned annually, announced 3 months in advance
- Minor releases (0.X.0) - Released quarterly with new features
- Patch releases (0.0.X) - Released as needed for bug fixes
Pre-release Versions
We use these pre-release identifiers:
- alpha (X.Y.Z-alpha.N) - Early testing, may have bugs and incomplete features
- beta (X.Y.Z-beta.N) - Feature complete, undergoing testing
- rc (X.Y.Z-rc.N) - Release candidate, ready for production testing
Support Policy
- Current major version - Full support with new features, bug fixes, and security patches
- Previous major version - Security patches and critical bug fixes for 12 months
- Older versions - No support; please upgrade
Deprecation Policy
When we deprecate a feature:
- Mark it as deprecated in the release notes
- Add deprecation warnings in the code
- Maintain the feature for at least 2 minor versions
- Remove in the next major version
- Provide migration guide for alternatives
How to Read This Changelog
Change Categories
Changes are grouped into categories:
- Added - New features and capabilities
- Changed - Changes to existing functionality
- Deprecated - Features that will be removed in future versions
- Removed - Features that have been removed
- Fixed - Bug fixes
- Security - Security-related changes and fixes
Version Links
Each version number links to the GitHub release page with:
- Full release notes
- Binary downloads
- Source code archive
- Docker images
Issue References
Changes reference GitHub issues where applicable:
#123- Links to issue or pull request@username- Credits contributor
Contributing
Found a bug? Want to request a feature? Here's how:
- Check existing issues - Search GitHub Issues
- Create an issue - Use our issue templates for bugs or features
- Submit a PR - See our Contributing Guide
Changelog Contributions
When submitting a PR:
- Add an entry to the
[Unreleased]section - Use the appropriate category (Added, Changed, etc.)
- Describe the change from a user's perspective
- Reference the issue or PR number
Example:
1
2
### Added
- Support for custom identity providers (#123) @contributor-name
Stay Updated
- Watch releases - Click "Watch" → "Releases only" on GitHub
- Subscribe to announcements - Join our mailing list
- Follow on Twitter - @agentweave
- RSS feed - Subscribe to our release feed
Archive
For a complete list of all releases, see the Releases page on GitHub.
Questions? See our Contributing Guide or ask in GitHub Discussions.