Skill UI

This guide details advanced architectural patterns for building complex AI systems using multiple Language Model instances (agents). It addresses the single-agent context bottleneck by distributing work, enabling parallelism, and facilitating specialization across different subtasks. Key patterns include Supervisor/Orchestrator, Peer-to-Peer/Swarm, and Hierarchical designs, providing guidelines for context isolation and effective coordination protocols.

Enables Claude to orchestrate multi-agent systems using AI SDK v5, setting up agents, handoffs, intelligent routing, and workflows across OpenAI, Anthropic, and Google when users request complex agent coordination.

Post and read messages on the AgentHub coordination board for competing agents, with channels for dispatch, progress, and results.

Dispatch dedicated agents with isolated contexts to handle multiple independent failures or tasks in parallel. Use when you have several unrelated issues that can be investigated separately without shared state, while preserving your own context for coordination and review.

Lambda is a compact, machine-native communication protocol specifically designed for agent-to-agent (A2A) messaging. It functions as a shared, unambiguous vocabulary built from 'atoms' (2-character codes), completely bypassing the need for human natural language interpretation. It is ideal for orchestrators, task delegation, and logging structured coordination signals (like heartbeats or state changes) in multi-agent systems. Crucially, Lambda is intended only for machine communication, not for user-facing dialogue.

This skill is designed for complex, multi-step tasks requiring coordination among multiple AI agents. It implements a robust workflow engine that first generates a detailed plan (task graph, topology) and then orchestrates parallel execution with adversarial verification. Use this when the task spans many files, requires multiple specialized roles, or involves an overall 'workflow' concept, ensuring structured, reliable, and auditable results.

A comprehensive guide designed to help researchers determine the optimal conference or journal venue (e.g., AAMAS, NeurIPS, ICML, EC) for multi-agent AI research. It emphasizes that the contribution must center on the interaction result—such as mechanism design, coordination, or negotiation—rather than simply applying a single-agent method in a multiagent testbed. This tool sharpens the research framing before writing begins.

This meta-skill orchestrates all agents within the ecosystem, providing automatic skill scanning, capability matching, multi-skill workflow coordination, and registry management. It helps manage complex tasks by coordinating multiple AI tools without executing them directly, ensuring efficient resource allocation and task planning.

This skill offers a design and workflow guide for intelligent context management within AI workflows. It focuses on capturing project state, enabling semantic retrieval, and supporting multi-agent coordination. Key features include state serialization, compression techniques, vector database integration, and knowledge graph construction. It helps preserve architectural decisions and facilitate knowledge transfer without implementing actual storage functions, serving as a strategic planning tool for context engineering.

Lambda is a compact, machine-native communication protocol specifically designed for agent-to-agent (A2A) messaging. It functions as a shared, unambiguous vocabulary built from 'atoms' (2-character codes), completely bypassing the need for human natural language interpretation. It is ideal for orchestrators, task delegation, and logging structured coordination signals (like heartbeats or state changes) in multi-agent systems. Crucially, Lambda is intended only for machine communication, not for user-facing dialogue.

A quota-aware parallel subagent coordination skill for Antigravity 2.0. Decomposes large multi-file tasks into isolated agent missions, manages model selection based on cost, and ensures efficient parallel execution while monitoring quota usage. Ideal for complex projects spanning multiple files and components.

This skill coordinates work across a RuFlo AI-agent swarm by discovering channels, reading signed updates, and checking claims. It helps manage multi-agent status, handoffs, ownership conflicts, and shared-work planning. Users must treat relay-sourced messages as untrusted and avoid irreversible actions unless explicitly requested. It ensures safe coordination without unauthorized state changes.