Hey @anflsc2024,

We have created an internal ticket to look into this which we will be tracking as "GHC-1267"

"output":
"Agent stopped due to max iterations."

I found the reason. I think the max iterations of 6 exactly of one act I loop through

Hey @anflsc2024

That would do it, so as you are hitting the max iterations I don't think there is a bug here. Let me know if you believe this to be a bug still and we can work out what to do next.

I still think there is a bug. Only the first item from the loop is processed in a continuous loop. I only got it working now by placing the agent in a sub-workflow. This indicates to me that there's something wrong with the main implementation.

Hey @anflsc2024,

Can you clarify what the bug is and provide a minimal workflow that reproduces it? to me it looks like it is working as expected if the Agent node is hitting a limit and not outputting an item the loop will finish this also happens with normal nodes so we have the option to continue on error and always output data so the output can be checked and passed back to the loop for the next run.

Bug.json

Here you can import the workflow

I encountered the same problem, occasionally. During the loop, a node gets the data of the previous loop.

Hey @anflsc2024,

Looking at your example workflow in the ScenePrompt code node you are using $input.first() which will always use the first input item, This is resulting in the same prompt being returning 15 times which is why it is using the same data.

By changing your code node to use the below it is running loops as expected.

// Process each input item individually and return a prompt for each
// Using itemsArray to ensure we preserve the original order
const result = [];

// Loop through each input item in the original order
for (const item of $input.all()) {
  // Create a prompt for this item
  result.push({
    Prompt: `You are an experienced anime storyboard artist and director for a detective anime series. Analyze the following scene and determine the optimal number and type of key frames needed to visually tell this story.
    
    SCENE: ${item.json.Scene || ''}
    Prompt: ${item.json.prompt || ''}
    Act: ${item.json.Act || ''}
    
    Your Task is to determine how many key frames are optimal for this scene. For EACH frame, create a detailed image generation prompt an use the **WriteImagePrompt Tool** with the following format:
    - Scene: [Act, scene number, frame number: short frame title]
    - ImagePrompt: [the image prompt]
    Create the prompts to give an image generator precise instructions on exactly how the image should look. Each shot should capture an important moment in the scene, and together they should cover the entire narrative development of the scene.`
  });
}

// Return all the processed items
return result;

@gavinkwang if you have a way to reproduce this issue it would be great to see it in in a workflow form.

@anflsc2024 after some more digging the team has managed to find an issue while looking at this, When using Tools it looks like the items are not correctly looped.