skills.rest
oddship
oddshipOFFICIAL
Explore all skills by this author →
WebsiteGitHub

meta-agent-creator

Generate markdown agent definitions with YAML frontmatter, model tiers, and tool permissions.

3|Updated Feb 24, 2026
Search Tags:#developer tools#ai development#configuration#agent creation#meta-programming#pi-agents
One-click install
npx skills add https://github.com/oddship/bosun --skill meta-agent-creator-oddship
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: meta-agent-creator
Source: https://github.com/oddship/bosun/tree/main/packages/pi-agents/skills/meta-agent-creator
Command: npx skills add https://github.com/oddship/bosun --skill meta-agent-creator-oddship

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the creation of specialized AI agents by providing a structured way to define their purpose, capabilities, and access permissions, ensuring they are well-equipped for specific tasks.

Core Features & Use Cases

  • Agent Definition: Generates .md files for new agents with correct YAML frontmatter.
  • Model Tier Configuration: Helps select appropriate AI model tiers (lite, medium, high, oracle) based on task complexity.
  • Tool & Extension Management: Defines necessary tools (read, grep, bash, etc.) and extensions for spawned agents.
  • Use Case: You need to create a dedicated agent for code review. This Skill helps you define its name, description, assign it the 'medium' model tier, grant it 'read' and 'grep' tools, and specify necessary extensions like 'pi-agents' and 'pi-question'.

Quick Start

Use the meta-agent-creator skill to create a new agent definition for code review.

Frequently Asked Questions about meta-agent-creator

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

FAQPage Schema
How do I create a specialized AI agent for code review?▼

To create a specialized AI agent, this Skill generates markdown files with proper YAML frontmatter, allowing you to define the agent's purpose, model tier, and specific tool permissions like read and grep.

What model tiers can I configure when building AI agents?▼

When building AI agents, you can configure model tiers including lite, medium, high, and oracle, allowing you to select the appropriate AI model based on the specific task complexity.

How do I manage tool permissions for Pi agent definitions?▼

Managing tool permissions for Pi agent definitions involves specifying necessary tools like read, grep, and bash within the generated YAML frontmatter to ensure spawned agents operate securely.

What is the best way to define access permissions for security analysis agents?▼

The best way to define access permissions is by generating a structured markdown file that configures appropriate extensions and tool permissions, ensuring the security analysis agent operates effectively within the Pi ecosystem.

Do I need specific extensions to run documentation agents in the Pi ecosystem?▼

Yes, documentation agents require specific extensions like pi-agents and pi-question to operate effectively within the Pi ecosystem, which are configured automatically when generating the agent definition.

Related Skills

Explore More Software Engineering Skills→
MadAppGangMadAppGang
7

task-complexity-router

Route tasks to AI model tiers based on complexity using keyword heuristics.

Official
Advanced
zpyoungzpyoung
4

pi-dev

Dispatch coding tasks to the pi coding agent via pi-watch.

Community
Intermediate
emptistemptist

pi-platform

Build and manage Pi extensions, tools, hooks, and skills.

Community
Advanced
prulloacprulloac

custom-agent-creator

Generate Copilot .agent.md or OpenCode JSON/Markdown agent configurations.

Community
Intermediate