xerxes-agent

Spawn and orchestrate multiple sub-agents within Xerxes for parallel task delegation.

27|2|Updated Jan 15, 2024
One-click install
npx skills add https://github.com/erfanzar/Xerxes-Agents --skill xerxes-agent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: xerxes-agent
Source: https://github.com/erfanzar/Xerxes-Agents/tree/main/src/python/xerxes/skills/xerxes-agent
Command: npx skills add https://github.com/erfanzar/Xerxes-Agents --skill xerxes-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinates and orchestrates multiple sub-agents inside Xerxes to tackle complex workloads, reducing manual coordination and unlocking parallel execution.

Core Features & Use Cases

  • Spawns and manages specialized sub-agents (coder, researcher, reviewer, planner, etc.) using AgentTool, SpawnAgents, TaskCreateTool, and SendMessageTool.
  • Enforces PARALLEL-IS-THE-DEFAULT workflows with guidance on worktree isolation for parallel code work, task delegation patterns, and result synthesis.
  • Supports end-to-end multi-agent projects by breaking work into focused agents, collecting results, and synthesizing final outputs for user delivery.
  • Includes best-practice guidelines for delegation patterns, resource budgeting, and failure handling to maintain robust operations.

Quick Start

Spawn three parallel coder agents to design and implement a backend feature and then review the combined result.

Frequently Asked Questions about xerxes-agent

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

FAQPage Schema
How do I orchestrate parallel multi-agent workflows for complex coding tasks?

You orchestrate parallel multi-agent workflows by spawning specialized sub-agents using tools like SpawnAgents and AgentTool. This delegates tasks such as coding, research, and testing to run in parallel, reducing manual coordination for complex workloads.

What is worktree isolation and when do I need it for multi-agent coordination?

Worktree isolation separates parallel code changes into distinct branches, preventing conflicts when multiple sub-agents code simultaneously. It is needed when running parallel-is-the-default workflows to ensure independent agents can work without overwriting each other's results.

Can I use multiple sub-agents for long-running research and testing projects?

Yes, you can spawn multiple specialized sub-agents like researchers and testers for long-running projects. The system breaks work into focused agents, collects their results, and synthesizes the final output for end-to-end project coordination.

What's the best way to delegate work across specialized agents in a development environment?

The best way to delegate work is using established delegation patterns with TaskCreateTool and SendMessageTool. This assigns focused tasks to specialized agents, applies resource budgeting, and handles failures to maintain robust parallel operations.

Are there limitations when spawning parallel agents for multi-team feature development?

Limitations include managing resource budgets and handling agent failures during parallel execution. The system provides guardrails and best-practice guidelines for failure handling and result synthesis to mitigate coordination issues in multi-team features.