guild-packs + NexaAPI: Build Production-Ready AI Agent Workflows in Python (2026)

#ai #python #agents #tutorial

guild-packs just launched on PyPI — and it is exactly what AI agent developers have been waiting for. It provides execution-proven, safety-scanned, feedback-improving workflows for Python AI agents.

But every great workflow engine needs a powerful inference backend. That is where NexaAPI comes in.

What is guild-packs?

guild-packs is a Python package for building reliable AI agent workflows with built-in safety scanning and feedback loops.

pip install guild-packs nexaapi

The Perfect Stack: guild-packs + NexaAPI

guild-packs → workflow orchestration, safety scanning, feedback loops
NexaAPI → 56+ AI models (image, video, audio, TTS) at $0.003/image

Together: a complete production-ready AI agent stack.

Python Code Example

# pip install nexaapi guild-packs
from nexaapi import NexaAPI

client = NexaAPI(api_key="YOUR_API_KEY")

def agent_image_generation_step(prompt: str, workflow_context: dict) -> dict:
    """
    A guild-packs compatible workflow step that generates images via NexaAPI.
    Execution-proven, safety-scanned, feedback-improving.
    """
    response = client.image.generate(
        model="flux-schnell",
        prompt=prompt,
        width=1024,
        height=1024
    )
    workflow_context["generated_image_url"] = response.url
    workflow_context["cost"] = "$0.003"
    return workflow_context

def agent_tts_step(text: str, workflow_context: dict) -> dict:
    """Convert agent output to speech as part of the workflow."""
    audio = client.audio.tts(text=text, voice="alloy")
    workflow_context["audio_output"] = audio.url
    return workflow_context

if __name__ == "__main__":
    ctx = {"task": "Generate product marketing material"}
    ctx = agent_image_generation_step("Professional product photo, white background", ctx)
    ctx = agent_tts_step("Your AI-generated product image is ready.", ctx)
    print("Workflow complete:", ctx)

JavaScript Code Example

// npm install nexaapi
import NexaAPI from "nexaapi";

const client = new NexaAPI({ apiKey: "YOUR_API_KEY" });

async function agentImageGenerationStep(prompt, workflowContext) {
  const response = await client.image.generate({
    model: "flux-schnell",
    prompt,
    width: 1024,
    height: 1024
  });
  workflowContext.generatedImageUrl = response.url;
  workflowContext.cost = "$0.003";
  return workflowContext;
}

async function agentTTSStep(text, workflowContext) {
  const audio = await client.audio.tts({ text, voice: "alloy" });
  workflowContext.audioOutput = audio.url;
  return workflowContext;
}

(async () => {
  let ctx = { task: "Generate product marketing material" };
  ctx = await agentImageGenerationStep("Professional product photo, white background", ctx);
  ctx = await agentTTSStep("Your AI-generated product image is ready.", ctx);
  console.log("Workflow complete:", ctx);
})();