agent-orchestration

Coordinate specialized agents with role-based dispatch and parallel worktrees.

8|2|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/FuZhiyu/superRA --skill agent-orchestration-fuzhiyu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-orchestration
Source: https://github.com/FuZhiyu/superRA/tree/main/skills/agent-orchestration
Command: npx skills add https://github.com/FuZhiyu/superRA --skill agent-orchestration-fuzhiyu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents agent-driven research tasks from becoming fragmented, context-limited, or insufficiently reviewed by coordinating specialized agents, isolated workspaces, task dependencies, and workflow state.

Core Features & Use Cases

  • Role-Based Dispatch: Assign implementer and reviewer roles with stage-specific objectives and appropriate model tiers.
  • Parallel Worktree Execution: Run independent tasks or reviews concurrently in isolated worktrees, then harvest changes safely.
  • Review and Recovery: Adjudicate reviewer findings, manage revision cycles, preserve task-tree state, and escalate material blockers.
  • Use Case: Coordinate parallel research or implementation tasks, require independent review before progression, and merge approved work into a coherent project state.

Quick Start

Use the agent-orchestration skill to dispatch the appropriate implementers and reviewers for the current task-tree frontier while preserving dependencies and workflow state.

Frequently Asked Questions about agent-orchestration

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I coordinate multiple agents for parallel research tasks without losing workflow state?

Coordinate specialized agents by applying role-based dispatch, isolated worktree execution for parallel work, and staged task dependencies to preserve reliable workflow state across complex research tasks.

What is agent orchestration for code review and research automation?

Agent orchestration assigns implementer and reviewer roles to specialized agents, managing revision cycles and escalating blockers to ensure independent review before progression in research automation workflows.

How do I run independent tasks concurrently in isolated worktrees and harvest changes safely?

Run independent tasks concurrently in parallel worktree isolation, then harvest changes safely by applying canonical implementer and reviewer contracts with explicit cleanup of temporary worktrees after approved commits.

Can I enforce task dependencies and require independent review before merging agent work?

Yes, you can enforce approved dependency sequencing and require independent reviewer adjudication before progression, ensuring all implementer commits pass review cycles prior to merging into coherent project state.

When should I not use agent role-based dispatch for implementation and review?

Avoid agent role-based dispatch when tasks lack clear dependency sequencing or independent review requirements, as the orchestration overhead of worktree isolation and reviewer adjudication outweighs benefits for simple linear tasks.

Does agent orchestration support escalating material blockers across superRA workflows?

Yes, agent orchestration supports escalating material blockers across superRA workflows by managing revision cycles, preserving task-tree state, and handing off unresolved issues to appropriate agents for resolution.