new-agent

Create swarm agent entries in swarm.json with security rules and prompt files.

6|Updated Mar 6, 2026

Search Tags:#delegation #orchestration #security-rules #prompt-guidance #add-agent #swarm.json

Ready to run

npx skills add https://github.com/jnotsknab/mux-swarm --skill new-agent-jnotsknab

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: new-agent
Source: https://github.com/jnotsknab/mux-swarm/tree/main/Skills/bundled-docker/new-agent
Command: npx skills add https://github.com/jnotsknab/mux-swarm --skill new-agent-jnotsknab

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill simplifies adding a new agent to the swarm by ensuring swarm.json entries are correctly structured, security rules are respected, and a prompt file is created according to guidelines.

Core Features & Use Cases

Structured swarm.json entry: Enforces the agent schema and required fields.
Security-guided onboarding: Applies mcpServers, canDelegate rules, and toolPattern guidance to prevent privilege escalation.
Prompt file generation: Creates the required prompt file at the specified promptPath and supports Sub-Agent Delegation when needed.
Guided workflow: Includes step-by-step steps and an example to illustrate adding a new agent.

Quick Start

Add a new agent entry to swarm.json following the field reference, then create the corresponding prompt file in the prompts directory using the specified promptPath.

Frequently Asked Questions about new-agent

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

FAQPage Schema

How do I add a new agent to my swarm configuration?▼

You add a new agent by updating the swarm.json file with a correctly structured entry, following the enforced schema, security rules, and required fields to ensure safe agent creation.

What security rules apply when onboarding a new swarm agent?▼

Security rules for onboarding a swarm agent involve configuring mcpServers, canDelegate rules, and toolPattern guidance to prevent privilege escalation within the swarm.json configuration.

How does sub-agent delegation work in swarm.json?▼

Sub-agent delegation in swarm.json is activated when the canDelegate field is true, which automatically generates delegation guidance within the associated agent prompt file.

How are prompt files generated for new swarm agents?▼

Prompt files for new swarm agents are generated at the specified promptPath directory, creating the required prompt file according to established guidelines and delegation rules.

What is the required schema structure for a swarm.json agent entry?▼

The required schema for a swarm.json agent entry enforces strict field definitions and documents mandatory fields to ensure correct configuration and safe agent onboarding.

Can I use this skill to update existing agents in swarm.json?▼

No, this skill facilitates the safe creation of a new swarm agent by defining the schema, enforcing security rules, and generating the prompt file, rather than updating existing configurations.