ax-playbook

Generate Ax playbook code for evolving and applying context knowledge in TypeScript.

2.9k|186|Updated Feb 23, 2023

Search Tags:#typescript #playbook #agent #update #context-engineering #evolve #ax

Ready to run

npx skills add https://github.com/ax-llm/ax --skill ax-playbook

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ax-playbook
Source: https://github.com/ax-llm/ax/tree/main/website/static/typescript/.well-known/agent-skills/ax-playbook
Command: npx skills add https://github.com/ax-llm/ax --skill ax-playbook

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you generate correct Ax playbook code for growing, applying, and persisting evolving task knowledge around typed programs and agents.

Core Features & Use Cases

Creates playbooks with playbook(program, { studentAI, teacherAI? }) for AxGen programs.
Supports offline growth with evolve(), online feedback with update(), and application with applyTo().
Handles agent playbooks, JSON snapshots, restoration, and context injection for production workflows.
Use it when you need to encode reusable context lessons, adapt prompts over time, or tune actor and responder stages.

Quick Start

Ask for a playbook implementation that matches your Ax program signature and specifies whether you want offline evolve or online update behavior.

Frequently Asked Questions about ax-playbook

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

FAQPage Schema

How do I create an evolving playbook for an AxGen program in TypeScript?▼

To create an evolving playbook for an AxGen program, use the playbook() function by passing your typed program signature and studentAI handle, optionally including a teacherAI for guided offline evolution and online updates.

What is context engineering for AI agents?▼

Context engineering for AI agents involves generating and applying evolving task knowledge to typed programs. It uses playbooks to encode reusable context lessons, adapt prompts over time, and tune actor and responder stages.

How do I persist and restore agent playbook context using JSON?▼

You persist and restore agent playbook context by saving JSON snapshots of the generated AxPlaybook state. You then restore these snapshots to inject accumulated context lessons back into your bound program.

Does the Ax playbook support both offline evolve and online update workflows?▼

Yes, the Ax playbook supports both offline growth using the evolve() function and online feedback adaptation using the update() function, allowing you to tune actor and responder stages by applying the resulting playbook back to the bound program.

When do I need a teacher AI handle for context playbooks?▼

You need a teacher AI handle for context playbooks when performing offline evolve() workflows to guide the student AI. It is optional for basic playbook generation and online update() feedback loops.