ax

Develop LLM applications with typed signatures using the @ax-llm/ax TypeScript library.

Updated Feb 15, 2026
One-click install
npx skills add https://github.com/eycjur/wandb_agent_hackathon --skill ax-eycjur
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ax
Source: https://github.com/eycjur/wandb_agent_hackathon/tree/main/llm-as-a-judge-mvp/.claude/skills/ax
Command: npx skills add https://github.com/eycjur/wandb_agent_hackathon --skill ax-eycjur

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill simplifies the development of LLM-powered applications by providing a type-safe framework for defining inputs, outputs, and interactions with various AI models.

Core Features & Use Cases

  • Type-Safe Signatures: Define clear input/output schemas for LLM calls using a fluent TypeScript API.
  • Multi-Provider Support: Easily switch between different LLM providers (OpenAI, Anthropic, Gemini, etc.).
  • Agents & Workflows: Build complex agents with tools and orchestrate multi-step AI workflows.
  • Use Case: Develop a customer support chatbot that can understand user queries, search a knowledge base, and generate helpful responses, all while maintaining type safety and allowing for easy provider switching.

Quick Start

Use the ax skill to create a simple generator that takes a question and returns an answer.

Frequently Asked Questions about ax

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

FAQPage Schema
How do I build type-safe signatures for LLM applications in TypeScript?

You can build type-safe signatures using a fluent TypeScript API to define clear input and output schemas for LLM calls. This ensures structured, maintainable, and robust interactions with various AI providers.

Can I switch between different LLM providers like OpenAI and Anthropic in the same workflow?

Yes, you can easily switch between different LLM providers like OpenAI, Anthropic, and Gemini. The framework supports multi-provider integration, allowing flexible model selection within your workflows.

What is the best way to orchestrate multi-step AI agents with tools?

Orchestrating multi-step AI agents is done by using a framework that supports building complex agents with tools and orchestrating workflows. This enables structured task execution and maintainable type safety across interactions.

Does this framework support building a customer support chatbot with a knowledge base?

Yes, the framework supports developing a customer support chatbot that can understand user queries, search a knowledge base, and generate helpful responses. It maintains type safety while allowing for easy provider switching.

Why do I need type-safe signatures for LLM workflow orchestration?

Type-safe signatures are needed for LLM workflow orchestration to provide structured, maintainable, and robust application development. They define clear input and output schemas, preventing runtime errors when integrating AI models.