🔀Tool Routing and Orchestration Skills
Browse skills in the Tool Routing and Orchestration category.
Sub Agent Patterns
A powerful skill for Claude agents.
Cloudflare Agents
A powerful skill for Claude agents.
Agents Sdk
A powerful skill for Claude agents.
Autonomous Agents
A powerful skill for Claude agents.
Ai Agents Architect
A powerful skill for Claude agents.
Behavioral Modes
A powerful skill for Claude agents.
Receiving Code Review
A powerful skill for Claude agents.
Dispatching Parallel Agents
A powerful skill for Claude agents.
Subagent Driven Development
A powerful skill for Claude agents.
pal-mcp-expert
Expert guidance for using the Pal MCP Server (zen-pal-nas). This skill should be used when working with multi-model AI orchestration, tool workflows (chat, thinkdeep, planner, consensus, debug, codereview, precommit, clink), configuration troubleshooting, or optimizing model selection strategies. Activates automatically when user mentions Pal MCP, zen-pal-nas, or specific tool names.
dispatching-parallel-agents
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
subagent
Execution skill for delegated tasks. Use this skill when you are a subagent that has been delegated a task by the tech lead. You must be diligent, hygienic, and deliver work that is "done" and "stable" according to the specification.
using-superpowers
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions
task-delegation
Project-agnostic skill for delegating execution work to subagents with skill activation, parallel execution, and quality verification.
coordinator-dx
Coordinator playbook for running multi‑repo, multi‑VM work in parallel without relying on humans copy/pasting long checklists.
task-router
Automatically analyze tasks and route to optimal agent(s). Use when receiving complex requests to determine which agents to spawn.
skill-selector
Self-driven skill selection system that uses internal curiosity asking (self-questioning) to analyze user requests, infer context from available information, and automatically recommend the most appropriate skills. Use when (1) User requests are ambiguous or unclear, (2) Multiple skills could potentially match a request, (3) User needs help discovering available skills, (4) Understanding user context is needed before recommending skills, or (5) User asks "what skills are available" or "which skill should I use".
subagent-creator
Create specialized Claude Code sub-agents with custom system prompts and tool configurations. Use when users ask to create a new sub-agent, custom agent, specialized assistant, or want to configure task-specific AI workflows for Claude Code.
claude-agent-sdk-ts
Build AI agents with Claude Agent SDK in TypeScript. Covers V1 query() API for batch workflows and V2 session API for interactive apps. Includes tool configuration, hooks, MCP servers, and multi-turn patterns.
parallel-dispatch
Coordinate simultaneous agent execution for independent tasks. Use when multiple tasks can run in parallel without dependencies.
multi-agent-dispatch
Dispatch tasks to remote VMs for parallel execution or specialized work.
roo-followup
Before calling roo_task, load this skill to handle Roo follow-ups. Use when coordinating Roo responses that return resultType/needsReply: ask the user with numbered options, summarize prior results, and re-send as a new Roo task (Roo cannot append in place).
dispatching-parallel-agents
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
mcp-integration
Configure and manage MCP (Model Context Protocol) servers for AI agent tooling.Use when adding MCP servers, configuring authentication (OAuth 2.1 or API keys),managing opencode.json, implementing token flows, or troubleshooting MCP connections.Covers registry patterns, PKCE authentication, and the Result-based service architecture.
hook-creator
Create and configure Claude Code hooks for customizing agent behavior. Use when the user wants to (1) create a new hook, (2) configure automatic formatting, logging, or notifications, (3) add file protection or custom permissions, (4) set up pre/post tool execution actions, or (5) asks about hook events like PreToolUse, PostToolUse, Notification, etc.
adk-orchestration
This skill should be used when the user asks about "multi-agent systems", "sub-agents", "delegation", "agent routing", "orchestration", "SequentialAgent", "ParallelAgent", "LoopAgent", "agent-to-agent", "A2A protocol", "agent hierarchy", "streaming", "real-time responses", "SSE", "server-sent events", "websocket", "bidirectional", "Live API", "voice", "audio", "video", "multimodal streaming", or needs guidance on building systems with multiple specialized agents working together or implementing real-time communication patterns.
cancel
Cancel active clarify loop
debate-multiverse
멀티 프로바이더 토론 - Claude/GPT/Gemini 3자가 각자의 관점에서 토론
unified-planning
Unified Planning Layer - 智能整合 spec-workflow 與 evolve PDCA
parallel-executor
Execute independent spec stages in parallel using subagents with real-time incremental apply. Use when spec has multiple stages without file conflicts. Automatically detects dependencies and runs maximum parallel agents.
agent-organizer
Assemble and coordinate multi-agent teams: decompose tasks, select skills/agents, design workflows, and manage execution with monitoring and recovery.
subagents-creator
Guide for defining and using Claude subagents effectively. Use when (1) creating new subagent types, (2) learning how to delegate work to specialized subagents, (3) improving subagent delegation prompts, (4) understanding subagent orchestration patterns, or (5) debugging ineffective subagent usage.
feature-orchestrator
This skill orchestrates complete feature development using specialized agents. Use when the user says "build feature", "implement feature", "develop feature end-to-end", "full feature lifecycle", "orchestrate development", or needs coordinated multi-agent feature work with discovery, design, implementation, and review phases.
technical-lead-role
Defines the technical lead role for main thread in agentic coding workflows - planning, delegation, coordination, and quality oversight
llm-application-patterns
This skill should be used when building production LLM applications in any language. It applies when implementing predictable AI features, creating structured interfaces for LLM operations, configuring language model providers, building agent systems with tools, optimizing prompts, or testing LLM-powered functionality. Covers language-agnostic patterns for type-safe contracts, modular composition, multi-provider support, and production deployment.
subagent-driven-development
Use when executing implementation plans with independent tasks in the current session
speckit-orchestrator
Intelligent workflow orchestrator that analyzes user requirements, recommends an appropriate workflow, and executes speckit agents in sequence.
long-session
Set up artifacts for multi-session autonomous work.Use when: (1) starting overnight/long autonomous runs,(2) user says "long session", "overnight", "autonomous",(3) work will span multiple sessions without user interaction.
agent-mail-guide
Guide for installing and using MCP Agent Mail across heterogeneous agent tools/VMs.
advanced-tool-usage
Guidelines for multi-stage tool orchestration and handling large data using 'redirect_tool_call'. Use this when you need to process large amounts of data without exhausting the context window or when building complex data pipelines.
getting-started
This skill is loaded automatically at session start via SessionStart hook. Establishes protocols for finding and using skills, checking skills before tasks, brainstorming before coding, and creating TodoWrite for checklists.
use-agents-dispatch
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
execute-approved-actions
Execute approved external actions via MCP servers in the Personal AI Employee system. This skill should be used when files appear in the /Approved folder (moved there by human after reviewing approval requests). It reads approval request files, invokes the appropriate MCP server tools (email, social media, browser automation), logs all actions to audit logs, and updates the system status. This skill implements the Human-in-the-Loop (HITL) execution phase for Silver tier. Use this skill whenever Claude Code detects new files in /Approved that are ready for execution.
beads
Multi-agent task coordination using Beads Village plugin tools. Use when work spans multiple sessions, has dependencies, needs file locking, or requires agent coordination. Covers claim/reserve/done cycle, dependency management, hierarchical decomposition, and session protocols.
chain-builder
Build and execute multi-step prompt chains for complex tasks
parallel
並列タスク実行ヘルパー。複数タスクの同時実行・結果統合
workflow
Use when executing multi-step processes requiring state persistence and agent coordination - defines two-tier protocol where subagents report PASS/FAIL status and main agent handles dispatch, advancement, and failure troubleshooting
worktree-swarm
Set up and orchestrate parallel Claude Code agents using git worktrees. Use this skill when the user wants to parallelize a feature across multiple agents, spawn worktrees, or merge agent work back together.
agent-builder-vercel-sdk
Build conversational AI agents using Vercel AI SDK + OpenRouter. Use when creating Next.js frontends with streaming UI, tool calling, and multi-provider support.
agent-select
Analyze tasks and recommend optimal sub-agent(s) for execution