This release adds 2 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Affected surfaces
Summary
AI summaryFixed token resource URIs, added logs://server URI scheme, and resolved database schema and authentication issues in the testing agent system.
Full changelog
Critical Bug Fixes
🔧 URI System Fixes
- Fixed token resource URIs: Corrected from
admintotoken://adminformat for proper MCP resource access - Server logs resource: Added proper
logs://serverURI scheme for debugging access - URI compatibility: All MCP resources now use consistent URI formatting
🧪 Testing Agent System Fixes
- Database schema errors: Fixed
agent_actionstable column reference issues preventing testing agent logging - Authentication issues: Resolved token authentication problems that prevented testing agents from accessing MCP tools
- Automatic validation: Testing agents now launch properly on task completion and perform comprehensive validation
✅ Verified Functionality
- Testing agent system fully operational with automatic launch workflow
- All URI formats working correctly across token, agent, tmux, and server log resources
- Authentication system stable with proper token handling
- Multiple testing agents can run simultaneously for concurrent validation
🚀 System Status
- Testing Agent Workflow: Task completion → Automatic testing agent launch → Comprehensive validation → Feedback to original agent
- URI Access: All @ mention resources accessible with proper formatting
- Database Operations: Stable with proper schema compatibility
- Multi-Agent Coordination: Enhanced with proper testing validation pipeline
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About rinadelph/Agent-MCP
A framework for creating multi-agent systems using MCP for coordinated AI collaboration, featuring task management, shared context, and RAG capabilities.
Related context
Beta — feedback welcome: [email protected]