ax-flow

Generate typed AxFlow workflow code for AI pipelines.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you write correct AxFlow workflows without wrestling with state wiring, execution order, or output typing mistakes.

Core Features & Use Cases

  • Typed workflow generation: Build flows with clear input and output contracts for multi-step AI pipelines.
  • Workflow orchestration: Create branching, looping, mapping, and parallel execution graphs with AxFlow.
  • Use Case: Use it when you need a DAG-style LLM pipeline that summarizes, classifies, routes, or combines multiple node outputs in one reliable program.

Quick Start

Ask me to generate an AxFlow workflow for your task, and include the inputs, outputs, and any branching or parallel steps you want.

Frequently Asked Questions about ax-flow

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

FAQPage Schema
How do I generate a typed AI pipeline with conditional routing and parallel execution in TypeScript?

You can generate a typed AI pipeline using AxFlow workflows to handle conditional routing, parallel execution, and map/reduce patterns. This approach uses node-before-execute ordering and flat state conventions to ensure reliable multi-node DAG pipelines.

What is the best way to structure a multi-node DAG workflow for LLM summarization and classification?

The best way to structure a multi-node DAG workflow is by using AxFlow orchestration with flat state conventions and .returns() output locking. This creates a reliable program that safely summarizes, classifies, or routes multiple node outputs.

How do I prevent state wiring and execution order mistakes when building branching LLM pipelines?

To prevent state wiring and execution order mistakes, generate AxFlow workflows that enforce node-before-execute ordering and safe auto-parallel execution rules. This ensures your branching LLM pipelines run without manual state wiring errors.

Does AxFlow support map/reduce patterns and loops for agent workflows?

Yes, AxFlow supports map/reduce patterns, loops, and agent workflows within typed AI pipelines. It applies safe auto-parallel execution rules to generate reliable workflow code for complex looping and branching structures.

Why do my TypeScript workflow outputs have incorrect typing during parallel execution?

Incorrect typing during parallel execution often stems from missing output locking. AxFlow uses .returns() output locking and flat state conventions to ensure typed AI pipelines maintain strict input and output contracts across parallel nodes.