config

Parse ax.yaml configuration, resolve paths, and load environment variables.

5|2|Updated Feb 8, 2026
One-click install
npx skills add https://github.com/project-ax/ax --skill config-project-ax
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: config
Source: https://github.com/project-ax/ax/tree/main/.claude/skills/ax/config
Command: npx skills add https://github.com/project-ax/ax --skill config-project-ax

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps manage the complex configuration, path resolution, and environment variable loading for the AX agent, ensuring a smooth and customizable operational setup.

Core Features & Use Cases

  • Configuration Management: Understand and modify ax.yaml settings for agent behavior, models, and providers.
  • Path Resolution: Navigate and manage the filesystem structure used by AX, including workspaces and agent directories.
  • Environment Variables & OAuth: Handle .env file loading and OAuth token refresh for secure and continuous operation.
  • Use Case: When setting up a new AX agent, you need to define which LLM models to use for different tasks and how to handle API credentials securely. This skill provides the necessary context and tools to configure these aspects.

Quick Start

Use the config skill to understand the structure of the ax.yaml file.

Frequently Asked Questions about config

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

FAQPage Schema
How do I validate configuration settings for an AI agent using Zod?

You validate AI agent configuration settings using Zod by parsing the `ax.yaml` file against a strict schema. This ensures your agent behavior, model selections, and provider details are correctly typed before execution.

How do I load environment variables and handle OAuth token refresh for automated agents?

Loading environment variables and handling OAuth token refresh involves processing `.env` files through a dedicated dotenv handler. This ensures your automated agents maintain secure and continuous API authentication during operation.

Can I route different LLM models to specific task types in a configuration file?

Yes, you can route different LLM models to specific task types by defining per-task-type model routing in your configuration file. This allows granular control over which models execute distinct agent actions.

What is the best way to manage filesystem paths and workspace directories for an agent?

Managing filesystem paths and workspace directories is best handled through centralized path resolution. This approach standardizes how agent directories and workspaces are accessed across your application, preventing broken file links.

Does this configuration approach support detailed history and scheduler settings?

Yes, this configuration approach supports detailed history and scheduler settings within the `ax.yaml` file. You can customize execution history retention and automated task scheduling to fit your operational requirements.

Why do I need a centralized configuration file for AI agent setup?

A centralized configuration file is needed for AI agent setup to consolidate path resolution, environment variables, and model routing. This creates a customizable and smooth operational foundation without scattering setup details.