skills.rest
project-ax
project-axOFFICIAL
Explore all skills by this author →
GitHub

ax-runners

Configure AX agent runners and select LLM transport options for sandbox execution.

5|2|Updated Feb 8, 2026
Search Tags:#ai-agent#llm-communication#runner-configuration#ax-agents#sandbox-execution
One-click install
npx skills add https://github.com/project-ax/ax-legacy --skill ax-runners
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ax-runners
Source: https://github.com/project-ax/ax-legacy/tree/main/.claude/skills/ax-runners
Command: npx skills add https://github.com/project-ax/ax-legacy --skill ax-runners

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill Unit assists in configuring and customizing the AX agents for different execution tasks while keeping the operation within a secure sandbox environment.

Core Features & Use Cases

  • Multi-Runner Setup: Select between pi-session (pi-coding-agent) and claude-code (Agent SDK) for running LLM communications.
  • Sandbox Execution: Facilitate agent tasks in a local subprocess sandbox with host audit.
  • Custom Tools & Features: Integrate custom tools and LLM transport selection.
  • Use Case: For software developers who need to tailor AI agent tasks within the source code repository, allowing for secure execution of tasks like code review, fetching web content, managing calendars, and maintaining user preferences.

Quick Start

Use the 'ax-runners' skill to configure an AI runner with a new set of tools or select an agent type for execution within your application.

Frequently Asked Questions about ax-runners

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

FAQPage Schema
How do I configure AI agents to run tasks within a sandbox?▼

To configure AI agents for sandbox execution, you can use the ax-runners skill to customize runner settings, register custom tools, and select different LLM communication transport options.

What transport options are available for LLM communication in sandbox execution?▼

Available LLM communication transport options include Python's IPC client and HTTP-based IPC, which are specifically utilized for facilitating kubernetes pod interactions during agent task execution.

Does sandbox execution support different types of AI agents?▼

Yes, sandbox execution supports multiple AI agent types, allowing you to select between pi-session for pi-coding-agent and claude-code for the Agent SDK depending on your specific task requirements.

Can I use custom tools with AI agents in a local subprocess sandbox?▼

Yes, you can integrate custom tools into the AI agent configuration to facilitate tasks within a local subprocess sandbox while maintaining host audit capabilities for secure execution.

What Python version is required for AI runner configuration?▼

AI runner configuration requires Python 3.8 or above to properly execute agent tasks and handle the underlying IPC mechanisms for LLM communication and data processing.

Why use a local subprocess sandbox for AI agent tasks?▼

A local subprocess sandbox provides a secure environment for executing AI agent tasks like code review and web content fetching, ensuring operations are isolated and audited on the host.

Related Skills

Explore More Software Engineering Skills→
sasha-computersasha-computer

debug

Analyzes agent/subprocess logs and adjusts configurations to resolve execution failures.

Community
Intermediate
Postpartum-genushyacinthus29Postpartum-genushyacinthus29
8

dotnet-managedcode-markitdown

Convert documents to deterministic Markdown in .NET applications.

Community
Intermediate
project-axproject-ax
5

runners

Document and manage agent runner implementations for pi-session and claude-code.

Official
Advanced
ricablericable
1

@neural-trader/mcp

Expose 87+ neural trading tools to AI agents via MCP.

Community
Advanced