ax-agent

Generate AxAgent code and runtime patterns for @ax-llm/ax.

2.9k|186|Updated Feb 23, 2023
One-click install
npx skills add https://github.com/ax-llm/ax --skill ax-agent-ax-llm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ax-agent
Source: https://github.com/ax-llm/ax/tree/main/website/static/typescript/.well-known/agent-skills/ax-agent
Command: npx skills add https://github.com/ax-llm/ax --skill ax-agent-ax-llm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps generate correct AxAgent code and runtime patterns for @ax-llm/ax, reducing errors when building tool-using assistants, child-agent hierarchies, and clarification flows.

Core Features & Use Cases

  • Minimal agent scaffolding: Produces small, modern agent definitions using the factory-style API.
  • Tool and namespace wiring: Handles namespaced functions, child agents, discovery mode, and runtime call-site conventions.
  • Execution safety and control: Covers clarification, finalization, bubble errors, state resume, and decision points for observability or memory-related workflows.
  • Use case: A developer needs a delegation-ready assistant with child agents, explicit tool namespaces, and a safe clarification protocol without hand-writing the runtime glue.

Quick Start

Ask for a minimal Ax agent implementation that matches your signature, tools, namespaces, and clarification behavior.

Frequently Asked Questions about ax-agent

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

FAQPage Schema
How do I create a TypeScript LLM agent with child agents and namespaced functions?

To create a TypeScript LLM agent with child agents and namespaced functions, you need correct agentIdentity definitions, contextFields, and protocol-aware runtime wiring. This generates valid, copyable AxAgent code using the factory-style API for delegation-ready assistants.

What is discovery mode in Ax agents and when should I use it?

Discovery mode in Ax agents is a runtime pattern for tool-using assistants that enables dynamic function resolution. Use discovery mode when building delegation-ready assistants requiring child-agent hierarchies and explicit tool namespaces without hand-writing the runtime glue.

How do I implement clarification flows and bubble-error handling in TypeScript LLM applications?

To implement clarification flows and bubble-error handling in TypeScript LLM applications, apply protocol-aware runtime wiring with decision points for observability. This approach covers state resume and finalization to ensure execution safety in tool-using assistants.

Does ax-agent work with the @ax-llm/ax framework for generating agent runtime patterns?

Yes, ax-agent generates correct AxAgent code and runtime patterns specifically for @ax-llm/ax. It applies to agent creation, child agents, namespaced functions, and clarification flows in TypeScript-first LLM applications, reducing errors when building tool-using assistants.

Why does my Ax agent runtime fail when wiring contextFields and agentIdentity?

Your Ax agent runtime fails when contextFields and agentIdentity are not correctly defined for protocol-aware runtime wiring. Generating valid AxAgent code requires precise factory-style API definitions to avoid errors in tool-using assistants and child-agent hierarchies.