Activity
Mon
Wed
Fri
Sun
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
What is this?
Less
More

Memberships

Maker School

2.6k members • $184/m

1Prompt License

244 members • Free

Creators Circle | Growth & AI

1.2k members • Free

AI Automation Agency Hub

250.5k members • Free

Vertical AI Builders

9.6k members • Free

AI Automation Society

147.6k members • Free

Accelerants

1.9k members • Free

AI Avengers

1.8k members • Free

4 contributions to AI Automation Society
Agent goes into infinite tool calling loop.
Hey everyone, I could really use some help with this issue I’ve been stuck on for the past few hours. I’m building a simple agent that: 1. Collects information from the user via chat 2. Passes that information to a sub-agent, which updates a CRM (HubSpot in this case) I decided to use a sub-agent approach because I plan to add more tools later, and I didn’t want the main prompt to become bloated. The idea was to keep things modular and dynamic. The problem:Whenever I provide the last required detail, the sub-agent successfully calls the HubSpot tool and gets an “ok” response. At that point, I’d expect the agent to stop and return the confirmation message back to me in the chat. Instead, it enters an infinite loop: - It keeps calling the same tool over and over, even though the first attempt already succeeded. - Eventually, it crashes with “agent stopped due to max iterations”. - Looking at the logs, I can see that the tool succeeds in the first iteration, and the main agent even produces a valid response that could have been returned to me—but instead, it just restarts the process and re-executes the tool. Question: Has anyone else run into this looping behavior with sub-agents and tool calls? Is there a known fix or a pattern I should be following to make sure the agent stops after a successful call? I’ve attached a snippet of the logs below for context. You can see the sub-agent succeeds on the first try, but the loop continues anyway. Any insights would be hugely appreciated,this is driving me nuts.
Agent goes into infinite tool calling loop.
1 like • Aug 26
@Learning To Change You basically need to structure the prompt better and really explain in it the tool flow and when it needs to be used.
1 like • Aug 27
@Learning To Change Yes! thank you
Need help using Switch module not passing data
Hey everyone, I need some help with the Switch node in n8n. I’m fairly experienced with the platform, but I'm running into some workflow issues that I haven’t been able to cleanly solve. Coming from Make (Integromat), I’m used to routers where data flows through and you can easily branch and iterate without losing access to previous steps. In n8n, when I use a Switch to branch the flow, I find that I can't reference earlier nodes unless I explicitly merge them back in downstream. That forces me to add Merge nodes just to pull in previous data, which clutters up the workflow and makes it hard to read. Is there a better way to handle this kind of branching and iteration in n8n? Something that doesn’t require wiring everything back together manually like this? Appreciate any insights.
Need help using Switch module not passing data
0 likes • Jul 18
@Shawn S Will try that out, I think it's look much better than having the merge. Thanks.
FOR HIRE: Free AI and Automation services
I'm looking to get on projects that will help your business become more efficient through AI and Automation. I am super fluent with Make.com, CRMs, and n8n. Check out my LinkedIn to see demo videos of my builds: www.linkedin.com/in/josue-hernandez04 You can reach out though LinkedIn or DM.
1 like • Apr 11
@Cenap Turgut sent you DM on LinkedIn
Journaling AI Agent
Your journal should be writing itself by now. Just like mine does. I built an AI agent that automates your daily journal entries. Just speak or text like you're chatting with a friend. Here’s how it works: → Voice note or text is sent via Telegram. → Whisper transcribes audio, or the text goes straight to a variable. → GPT-based agent matches your tone and organizes your thoughts. → A custom Notion agent structures and stores the entry, complete with mood rating, highlights, takeaways, and tomorrow’s to-do list. The results are clean, human-like, and personalized. It’s fast. It’s simple. It’s like having a personal journaling assistant on autopilot. See the full demo below to see in action. 🚀 https://www.linkedin.com/posts/josue-hernandez04_your-journal-should-be-writing-itself-by-activity-7312902712906440705-sTZd?utm_source=share&utm_medium=member_desktop&rcm=ACoAADoPOvIB-Tt89bgXL8I19NJfQhu1TvJNPmI
1
0
1-4 of 4
Josue Hernandez
3
45points to level up
@josue-hernandez-1795
Automation and AI Enthusiast

Active 14h ago
Joined Feb 12, 2025
Powered by