nexusos

Manage NexusOS multi-agent orchestration, dispatch tasks, and inspect logs.

1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/cryptopafi/nexusos-skills --skill nexusos
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nexusos
Source: https://github.com/cryptopafi/nexusos-skills/tree/main/nexusos
Command: npx skills add https://github.com/cryptopafi/nexusos-skills --skill nexusos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

NexusOS unifies control of a local multi-agent orchestration system so users can inspect system health, dispatch classified tasks, retrieve logs, and trigger heartbeats without manually exploring workspace files and registries. It removes the friction of routing tasks to the correct agent, estimating budget/complexity, and following safe dispatch and retry protocols.

Core Features & Use Cases

  • Status and Health: Aggregate GENIE, SENTINEL, and agent health files into a concise status report with colored indicators.
  • Smart Dispatch: Classify task descriptions, estimate complexity and budget, check agent registry status, and create tracked workspaces or perform in-process dispatch.
  • Logs & Troubleshooting: Tail per-agent heartbeat logs, run smoke tests, and present actionable errors when scripts or files are missing.
  • Task Lifecycle & Recovery: Create DISPATCH.md/PROGRESS.md workspaces, enforce judge rules for MEDIUM/HIGH tasks, and follow retry/escalation protocols on failure.

Quick Start

Dispatch a task by telling NexusOS a short description, for example: /nexusos dispatch Research competitor SMSads performance in Kenya to auto-classify, estimate budget, and route to the appropriate agent.

Frequently Asked Questions about nexusos

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

FAQPage Schema
How do I dispatch tasks to specific agents in a multi-agent orchestration system?

To dispatch tasks in multi-agent orchestration, provide a short task description to automatically classify complexity, estimate budget, check agent registry status, and route the work to the appropriate active agent.

How do I monitor agent health and retrieve heartbeat logs for local task automation?

Monitoring agent health involves aggregating GENIE, SENTINEL, and agent health files into a concise status report, while retrieving heartbeat logs requires tailing per-agent logs to troubleshoot local workflows.

Can I use local shell scripts and configuration files for multi-agent task routing?

Yes, multi-agent task routing operates by reading local ~/.nexus configuration and workspace files, invoking trusted shell scripts when available to query agent health and manage task dispatch safely.

What is the best way to handle task lifecycle recovery and retries in multi-agent workflows?

Task lifecycle recovery in multi-agent workflows involves creating DISPATCH.md and PROGRESS.md workspaces, enforcing judge rules for medium and high complexity tasks, and following automated retry and escalation protocols on failure.

Does multi-agent task dispatch require user confirmation for high-budget operations?

Yes, multi-agent task dispatch requires explicit user confirmation for high-budget operations, ensuring safe execution by estimating complexity and budget before creating tracked workspaces or performing in-process dispatch.

Why are my local orchestration smoke tests failing to find missing workspace files?

Local orchestration smoke tests fail when scripts or files are missing, presenting actionable errors that require verifying your local ~/.nexus configuration and workspace registries before retrying the operation.