agent-manager

Coordinate and monitor Claude Code sub-agents with a JSONL registry.

Updated Aug 23, 2026

Search Tags:#automation #dashboard #orchestration #registry #claude-code #task-tracking #agent-management

Ready to run

npx skills add https://github.com/leonardofhy/openclaw-workspace --skill agent-manager-leonardofhy

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-manager
Source: https://github.com/leonardofhy/openclaw-workspace/tree/main/skills/agent-manager
Command: npx skills add https://github.com/leonardofhy/openclaw-workspace --skill agent-manager-leonardofhy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill provides a centralized lifecycle management layer for Claude Code sub-agents, enabling spawn, status dashboards, history, and auto-cleanup across multiple concurrent tasks.

Core Features & Use Cases

Agent orchestration: spawn, track, and name sub-agents with task prompts.
Visibility: real-time status dashboard and history logs.
Resource hygiene: auto-cleanup of completed or stale sessions to conserve resources.

Quick Start

Spawn a new agent with a descriptive name and a task, then monitor its status in the dashboard.

Frequently Asked Questions about agent-manager

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

FAQPage Schema

How do I orchestrate multiple Claude Code sub-agents at the same time?▼

You can orchestrate multiple Claude Code sub-agents by spawning, naming, and tracking them with task prompts. This skill coordinates concurrent execution using a local JSONL registry and file locking to ensure safe, reliable registry updates.

Can I view the status of concurrent tasks spawned by Claude Code sub-agents?▼

Yes, you can view the status of concurrent tasks through a real-time status dashboard. It monitors active sub-agents using pid-based status checks and maintains a history log of completed sessions for full visibility.

How do I clean up stale or completed Claude Code sub-agent sessions?▼

You can clean up completed or stale Claude Code sub-agent sessions using the auto-cleanup feature. This resource hygiene function removes old sessions from the JSONL registry to conserve system resources.

Does this sub-agent orchestration tool support model aliasing?▼

Yes, this sub-agent orchestration tool supports model aliasing when spawning sub-agents. It integrates this functionality directly into the local registry to manage concurrent task execution effectively.

What is the best way to track concurrent task prompts for Claude Code agents?▼

The best way to track concurrent task prompts is by using a centralized lifecycle management layer. It logs agent history and monitors status via a dashboard, relying on a local JSONL registry for persistent tracking.