Design TX V4 meshes and agent workflows with SDK-based workers and SQLite queues.
npx skills add https://github.com/eighteyes/tx --skill mesh-builder
Please help me install this Agent Skill. Skill: mesh-builder Source: https://github.com/eighteyes/tx/tree/main/.claude/skills/mesh-builder Command: npx skills add https://github.com/eighteyes/tx --skill mesh-builder
This Skill provides a comprehensive blueprint for building TX V4 meshes and agents, covering SDK-based worker architecture, prompt templates, message protocols, HITL workflows, and testing patterns to accelerate multi-agent development.
Create a simple mesh using a minimal config and verify messages appear in .ai/tx/msgs/
High-intent search queries and answers about installing and using this skill.
TX V4 meshes define agent topology, routing, and coordination using SDK-based workers, message protocols, and SQLite queues. Start by specifying worker architecture, routing logic (sequential, parallel, bidirectional, iterative), and agent prompts in your mesh config, then verify messages flow through the .ai/tx/msgs/ directory.
TX V4 improves scalability and agent coordination through enhanced SDK integration, refined message protocols, and better HITL workflow support compared to V3. The V4 architecture enables more flexible routing scenarios and semantic model mapping across distributed workers.
Use the Claude Agent SDK to spawn workers that stream results into a SQLite-backed message queue. Configure message frontmatter schemas and routing rules to coordinate agent communication, then monitor .ai/tx/msgs/ to verify protocol compliance and message delivery.
Yes. TX V4 meshes support slash command routing for user-driven agent selection and semantic model mapping to resolve the correct agent based on task context. Define routing logic in your mesh config alongside model resolution criteria.
TX V4 provides test-oriented patterns covering HITL workflows, iterative agent loops, and multi-agent coordination scenarios. Use message queue inspection, frontmatter validation, and routing verification to ensure agents collaborate as expected.
No, but understanding agent architecture and message-driven coordination helps. TX V4 is designed for both newcomers and V3 users; the mesh-builder Skill covers foundational concepts through advanced routing and testing strategies.
terrene-foundationDeploy scalable Kailash infrastructure from SQLite to multi-worker using environment variables.
DatTran26Run N autonomous iterations optimizing a measurable metric with git history.
plurigridManage tmux sessions, windows, and panes across terminal workflows.
EarthmanWebOrchestrate cloud-based AI agent swarms for event-driven workflow automation.
