ax-python-ai

Create Python provider clients for axllm across OpenAI, Gemini, and Anthropic workflows.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you write Python code that creates and configures axllm provider clients without guessing the correct native call shape.

Core Features & Use Cases

  • Provider Client Setup: Build clients for OpenAI-compatible endpoints, OpenAI Responses, Gemini, Anthropic, routers, and balancers.
  • Deterministic Local Testing: Use scripted no-key transports for predictable examples and mapping checks.
  • Credentialed API Workflows: Generate real network calls only when explicit provider credentials are available.
  • Example-First Guidance: Start from generated package examples and native AxIR documentation before inventing new patterns.

Quick Start

Use the ax-python-ai skill to draft a Python snippet that creates an axllm client for the provider you need, using a real API key or a scripted no-key transport as appropriate.

Frequently Asked Questions about ax-python-ai

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

FAQPage Schema
How do I create a Python provider client for Anthropic or Gemini using axllm?

To create a Python provider client for Anthropic or Gemini, you use the ax-python-ai skill to generate deterministic Ax Python code based on native axllm APIs. It handles provider-specific configurations so you avoid guessing the correct native call shape.

Can I test axllm provider clients locally without providing real API keys?

Yes, you can test axllm provider clients locally without API keys by using scripted no-key transports. This feature provides deterministic local testing for predictable examples and mapping checks without requiring explicit provider credentials.

What is the best way to configure an OpenAI-compatible endpoint in axllm?

The best way to configure an OpenAI-compatible endpoint in axllm is by generating a Python snippet through the ax-python-ai skill. It uses example-first guidance from generated package examples and native AxIR documentation to establish the correct client setup.

Does the axllm Python code generation support router and balancer workflows?

Yes, the axllm Python code generation supports router and balancer workflows. The skill generates Ax Python code that covers client setup for routers and balancers alongside OpenAI Responses, Gemini, and Anthropic provider configurations.

When do I need explicit credentials for axllm API workflows in Python?

You need explicit credentials for axllm API workflows in Python when generating real network calls. The skill ensures credentialed API workflows execute actual requests only when explicit provider credentials are available, otherwise relying on scripted no-key transports.