TX Mesh Architect & Tester

소개

This skill provides a comprehensive framework for architecting and validating multi-agent systems, known as meshes, within the tx environment. It guides users through creating specialized agent configurations, defining reactive roles, and implementing robust end-to-end testing strategies. By leveraging a centralized event log architecture and tmux-based session validation, it ensures that complex AI workflows are reliable, auditable, and easily debuggable through immutable message tracking and dynamic routing tables.

주요 기능

• 2 GitHub stars
• Centralized Event Log Architecture for immutable, chronological message tracking
• Session-based E2E Validation using tmux state capture and file verification
• Offset Tracking to manage message consumption and prevent duplicate processing
• Human-In-The-Loop (HITL) workflow patterns for guided Q&A and approvals
• Dynamic Routing Tables to automate multi-agent coordination and message flow

사용 사례

• Debugging complex asynchronous agent communications via centralized message logs
• Implementing E2E tests for mesh deployments using the standardized E2EWorkflow class
• Architecting a multi-agent 'ask-answer' workflow with automated status routing