Build typed LLM applications with Ax in TypeScript.
npx skills add https://github.com/ax-llm/ax --skill ax-llm
Please help me install this Agent Skill. Skill: ax-llm Source: https://github.com/ax-llm/ax/tree/main/website/static/typescript/.well-known/agent-skills/ax-llm Command: npx skills add https://github.com/ax-llm/ax --skill ax-llm
This Skill helps developers build LLM applications with a single TypeScript-first model for typed generation, provider abstraction, agents, flows, and optimization workflows.
Ask for a concise TypeScript example that shows how to create an Ax generator, agent, or flow with @ax-llm/ax and explains the required imports and provider setup.
High-intent search queries and answers about installing and using this skill.
To build typed LLM applications in TypeScript, use Ax generators with string DSLs, fluent builders, or Standard Schema validators like Zod and Valibot to enforce structured outputs. This provides reliable type safety for your structured generation workflows.
Yes, you can integrate multiple AI providers like Anthropic, Gemini, OpenAI, Grok, and Mistral in one TypeScript project. Ax provides a multi-provider abstraction layer, allowing you to access OpenAI-compatible endpoints, Azure OpenAI, and others through a unified interface.
The best way to orchestrate LLM agents and workflows in TypeScript is using Ax flows. Flows support tool use, runtime execution, checkpoints, memory, delegation, branching, loops, and parallel orchestration, enabling complex workflow orchestration for multi-agent systems.
Yes, Ax supports cross-language runtime generation. You can define one typed Ax program in TypeScript and deploy the same semantic workflow across Python, Java, C++, Go, and Rust builds, ensuring consistent application logic across diverse runtime environments.
You optimize few-shot prompts and evaluate LLM workflows in TypeScript using Ax's optimization and evaluation features. This includes few-shot bootstrapping, GEPA, portable optimizer artifacts, and apply/evaluate workflows to refine and assess your LLM application performance.
jadecliGenerate TypeScript AI provider integration code using @ax-llm/ax.
ax-llmRefine Go Ax generation workflows with evaluator feedback and optimizer APIs.
ax-llmGenerate TypeScript provider setup and configuration code for @ax-llm/ax.
juanreSwitch between LLM providers and access raw SDKs via get_provider().
