Hi everyone! 
I’m running into a strange issue with my AI Agent setup. I’m trying to call the “image agent” tool multiple times in a row as part of a workflow. The first two calls work exactly as expected — the tool runs, and I get the correct processed result.
However, on the third run, the workflow doesn’t call the tool at all. Instead, it seems to hallucinate a reply — returning a fabricated JSON response that was never generated by the tool. The reply follows the formatting of my rules but contains fake data. As a result, the image processing step is skipped entirely.
My Setup
• Image Workflow → triggers
• Image Agent Coordinator → formats the prompt and passes to image agent
• Image Agent → selects and runs a specific image processing tool
(e.g. removeBackground, createPoster, scaleImage, etc.)
The issue happens in the Image Agent Coordinator, which stops calling the tool starting with the 3rd execution.
AI Agent Rules (Coordinator)
1. Role
You are the Image Agent Workflow Coordinator.
Your primary responsibility is to:
• Receive a prompt and url
• Pass this input directly to the Image Agent
• Wait for its full JSON response
• Return that exact response to the caller (e.g. webhook) without alteration
You do not decide which tool to use. You are not responsible for interpreting the prompt or executing image processing tools. That is the role of the Image Agent.
2. Input
You receive 3 fields:
• url: the image URL to process
• prompt: a textual description of the user’s request
• sessionId
3. Flow of Execution
1. Pass url and prompt to the Image Agent.
2. Wait for the Image Agent’s full response.
3. Return exactly the same JSON response without changing, omitting, or reordering any fields.
4. JSON Format Enforcement
You must return:
• A valid JSON object
• In the exact structure provided by the Image Agent
• Without adding or removing any fields
Do not:
• Reformat field names
• Change null to empty strings or vice versa
• Reorder fields
• Add metadata
• Strip any fields
5. Error Propagation
If the Image Agent returns an error, you must still return the JSON as-is.
6. Example
Prompt: “Remove the background from this image”
URL: “https://cdn.example.com/image.jpg”
✅ You send both to the Image Agent.
✅ Image Agent returns:
{
"status": "success",
"action": "removeBackground",
"link": "https://cdn.example.com/image-clean.png",
"text": "Background removed successfully."
}
✅ You return exactly that.
✅ Final Reminder:
Always call the Image Agent tool.
Do not retry or loop.
No caching. No assumptions. No interpretation.
What I Observed
I’m using the gpt-4o-mini model
Each request has a new image URL and prompt
Only the third call fails, and it’s consistent
The third output looks something like this
{
“status”: “success”,
“action”: “createPoster”,
“link”: “https://”,
“text”: “Poster created successfully.”
}
…but the tool was never called — this is fabricated by the AI.
Questions
• Why might the agent skip the tool call only on the third run?
• Is it a model behavior quirk, internal state issue, or something wrong with how I set up the rules?
• Any known issues with GPT-4o-mini skipping tool calls or hallucinating outputs?
• Would love to hear if anyone has faced this or has a workaround 
Information on your n8n setup
- n8n version: 1.85.4
- Database (default: SQLite):
- n8n EXECUTIONS_PROCESS setting (default: own, main):
- Running n8n via : web app
- Operating system: macOS Sequoia 15.0
)