manager

Route AI tasks across backends with parallel execution and retries in MCP workflows.

Updated Aug 23, 2026

Search Tags:#automation #workflow #orchestration #task routing #coordinator #multi-backend

Ready to run

npx skills add https://github.com/AIWander/manager-universal --skill manager-aiwander

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: manager
Source: https://github.com/AIWander/manager-universal/tree/main/skills/manager
Command: npx skills add https://github.com/AIWander/manager-universal --skill manager-aiwander

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables automated coordination and management of AI tasks across multiple backends within MCP, streamlining workflow execution and task delegation.

Core Features & Use Cases

Task Delegation and Routing: Automatically determine and assign tasks to the most appropriate backend, such as Claude Code, Codex, Gemini, or GPT, based on task nature and success history.
Workflow Automation: Facilitate multi-step, multi-file AI operations with parallel execution, retries, and rollbacks for robust task management.
Use Case: Simplify complex software refactoring workflows by decomposing into subtasks, delegating to specialized AI backends, and coordinating results for deployment.

Quick Start

Use the manager skill to delegate a multi-file code refactoring task and monitor progress until completion.

Frequently Asked Questions about manager

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

FAQPage Schema

How does multi-backend task routing work for MCP workflows?▼

Multi-backend task routing for MCP workflows works by automatically evaluating task requirements and assigning operations to the most appropriate AI backend, such as Claude Code, Codex, Gemini, or GPT, based on task nature and historical success rates to optimize execution outcomes.

Can I run parallel execution and retries for multi-file AI operations?▼

Yes, you can run parallel execution and retries for multi-file AI operations. The orchestration process supports multi-step workflows with built-in retry logic and rollback capabilities to ensure robust task durability and automated recovery during complex software refactoring.

What is the best way to coordinate AI tasks across different backends?▼

The best way to coordinate AI tasks across different backends is to use an orchestration layer that handles task delegation, monitors execution progress, and aggregates results. This streamlines workflow execution by decomposing complex tasks into specialized subtasks for targeted backends.

Do I need any specific dependencies to automate task delegation in MCP?▼

No specific dependencies are required to automate task delegation in MCP. The orchestration operates independently to route and manage tasks across available AI backends, allowing you to initiate multi-file refactoring and execution monitoring without installing external packages.

When should I use workflow orchestration for AI task management?▼

You should use workflow orchestration for AI task management when executing complex, multi-step software refactoring that requires coordinating multiple specialized backends. It is essential when tasks need parallel execution, automated retries, and rollbacks to guarantee durable and successful completion.