managing-agents

Manage orchestration agent lifecycles from creation to cleanup.

Updated Nov 24, 2025
One-click install
npx skills add https://github.com/Sixallfaces/orkestr --skill managing-agents-sixallfaces
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: managing-agents
Source: https://github.com/Sixallfaces/orkestr/tree/main/skills/managing-agents
Command: npx skills add https://github.com/Sixallfaces/orkestr --skill managing-agents-sixallfaces

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Users need to create specialized AI agents for unique tasks but require a robust system to manage their lifecycle, from temporary use to permanent promotion and proper organization. This Skill provides comprehensive tools for agent creation, namespacing, promotion, and cleanup, ensuring a clean and efficient agent ecosystem.

Core Features & Use Cases

  • Temp Agent Lifecycle Management: Create ephemeral agents for single workflows, use them, and then decide to promote them to permanent status or automatically clean them up.
  • Agent Promotion: Easily convert successful temporary agents into permanent, reusable defined agents, making them discoverable and available across all future workflows.
  • Namespacing & Conflict Resolution: Automatically handle agent namespacing (orchestration:) to prevent conflicts with built-in agents and ensure proper routing.
  • Agent Maintenance: Update, delete, and track usage statistics for your defined agents, keeping your agent library optimized and relevant.
  • Use Case: After a workflow successfully uses a temporary security-scanner agent to find critical vulnerabilities, promote it to a permanent orchestration:security-scanner agent for reuse in all future CI/CD pipelines.

Quick Start

I want to create a custom agent for security auditing. How do I manage its lifecycle?

Frequently Asked Questions about managing-agents

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

FAQPage Schema
How do I create and manage temporary agents in my workflow?

Temporary agents are ephemeral AI agents created for single workflows under temp-agents/. Create them during workflow design, use them within task execution, then decide whether to promote them to permanent status or let them auto-clean. This lets you test agent patterns without cluttering your permanent agent registry.

What's the best way to convert a temporary agent into a permanent one?

Agent promotion moves successful temporary agents from temp-agents/ into the agents/ directory with the orchestration: namespace prefix, making them reusable across all future workflows. After testing a temp-agent's effectiveness, promotion registers it permanently in your agent registry for discovery and consistent routing.

How does namespacing prevent conflicts between built-in and custom agents?

Namespacing automatically prefixes custom agents with orchestration: to separate them from built-in agents, ensuring deterministic routing and conflict resolution during workflow execution. This convention keeps your agent namespace clean and prevents name collisions when invoking agents via Task or flow syntax.

Can I update or delete agents after they're promoted to permanent status?

Yes. Agent maintenance lets you update, delete, and track usage statistics for defined agents in agents/. This keeps your permanent agent library optimized, removing unused agents and refining active ones based on performance metrics.

When should I use temporary agents versus defined permanent agents?

Use temporary agents to experiment with new agent designs or one-off tasks without commitment. Once an agent pattern proves valuable and reusable across multiple workflows, promote it to permanent status for discovery, versioning, and lifecycle management across your orchestration ecosystem.

How are agents resolved and routed during workflow execution?

Agent resolution applies deterministic routing rules to match agent invocations to their definitions. Built-in, plugin-defined, and orchestration:-namespaced agents are resolved based on defined-time definitions and temp-agent file locations, ensuring consistent agent selection and execution within workflows.