agents

Spawn, monitor, and terminate autonomous agents via the gobby-agents MCP server.

19|3|Updated Dec 15, 2025
One-click install
npx skills add https://github.com/GobbyAI/gobby --skill agents-gobbyai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agents
Source: https://github.com/GobbyAI/gobby/tree/main/src/gobby/install/shared/skills/agents
Command: npx skills add https://github.com/GobbyAI/gobby --skill agents-gobbyai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill manages subagent spawning via the gobby-agents MCP server, enabling spawn, cancel, list, and result retrieval for autonomous agents within a session-driven workflow.

Core Features & Use Cases

  • Spawn named or ad-hoc agents with preconfigured workflows
  • Track, list, and monitor agent runs across sessions
  • Retrieve results and gracefully stop or kill agents when needed
  • Support orchestrator and worker workflows for coordinated task execution

Quick Start

Use the agents subcommand to spawn a named agent, for example: /gobby agents meeseeks-claude session_task=#123

Frequently Asked Questions about agents

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

FAQPage Schema
How do I spawn autonomous AI agents within a workflow?

To spawn autonomous AI agents within a workflow, use the agents subcommand to trigger named or ad-hoc agents with preconfigured workflows, such as passing a session task identifier to initiate the run.

Can I track and monitor active agent runs across sessions?

You can track and monitor active agent runs across sessions by using the list_agents tool to view running agents and retrieve their status throughout the orchestration lifecycle.

What is the best way to stop or terminate a running autonomous agent?

The best way to stop a running autonomous agent is by using the stop_agent tool for graceful termination or the kill_agent tool to immediately terminate the agent run.

Does the agents MCP server support orchestrator and worker workflows?

The agents MCP server supports orchestrator and worker workflows for coordinated task execution, allowing you to manage complex multi-agent runs and retrieve individual results.

How do I retrieve results from a spawned subagent?

To retrieve results from a spawned subagent, use the get_agent_result tool after the autonomous agent completes its assigned workflow task within the session.

When should I use kill_agent instead of stop_agent for agent lifecycle management?

Use stop_agent for graceful agent lifecycle management when tasks can finish safely, and use kill_agent when you need to immediately force-terminate an autonomous agent run.