agents-manager

Manage AI agent lifecycles, routing, and permissions in the Ravi runtime.

6|2|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/filipexyz/ravi --skill agents-manager-filipexyz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agents-manager
Source: https://github.com/filipexyz/ravi/tree/main/src/plugins/internal/ravi-system/skills/agents
Command: npx skills add https://github.com/filipexyz/ravi --skill agents-manager-filipexyz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the complexity of managing long-lived AI agent identities, their specific runtime configurations, and their integration into communication channels like WhatsApp without requiring system restarts.

Core Features & Use Cases

  • Agent Lifecycle Management: Create, configure, and delete operational agent identities with custom runtimes and models.
  • Dynamic Routing: Route messages from specific WhatsApp groups or contacts to designated agents based on flexible patterns.
  • Permission Control: Manage agent capabilities, tool access, and Bash execution rights through a centralized permission provider runtime.

Quick Start

Use the agents-manager skill to create a new agent named support-bot in the directory ~/ravi/support with the codex provider and gpt-5.5 model.

Frequently Asked Questions about agents-manager

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

FAQPage Schema
Does the agents-manager skill support real-time configuration updates for autonomous agents?

Yes, the agents-manager skill supports real-time configuration updates for autonomous agents. It allows you to dynamically manage runtime providers, models, and communication channel associations without requiring any system restarts.

Can I route WhatsApp messages to specific AI agents based on contact or group patterns?

Yes, you can route WhatsApp messages to specific AI agents through dynamic routing. This feature directs messages from designated contacts or WhatsApp groups to the appropriate autonomous agent based on flexible, customizable routing patterns.

How do I control permissions and tool access for autonomous AI agents?

You control permissions for autonomous AI agents through a centralized permission provider runtime. This mechanism manages agent capabilities, tool access, and Bash execution rights securely by assigning granular permissions to each operational identity.

What is the best way to orchestrate long-lived AI agent identities in a communication channel?

The best way to orchestrate long-lived AI agent identities is to use a management skill that associates operational identities with specific communication channels and runtime providers. This ensures secure execution and seamless integration without downtime.

Does the agents-manager skill support real-time configuration updates for autonomous agents?

Yes, the agents-manager skill supports real-time configuration updates for autonomous agents. It allows you to dynamically manage runtime providers, models, and communication channel associations without requiring any system restarts.