Skip to content
AI tool calling con...
 
Notifications
Clear all

AI tool calling consistently returns the same results despite varying inputs

13 Posts
8 Users
0 Reactions
3 Views
davidpp
(@davidpp)
Posts: 3
Active Member
Topic starter
 

I've encountered a problem with callin.io and tool calling when using AI agents.

When the same tool is invoked multiple times for a single user input, the tool's response remains consistent, irrespective of the input provided.

Call 1

Call 2

I've experienced the same issue with other tools, such as Jira.

Currently using Version 1.66.0 (self-hosted).

 
Posted : 14/11/2024 2:46 pm
n8n
 n8n
(@n8n)
Posts: 97
Trusted Member
 

It appears your topic is missing some crucial details. Could you please provide the following information, if relevant?

  • callin.io version:
  • Database (default: SQLite):
  • callin.io EXECUTIONS_PROCESS setting (default: own, main):
  • Running callin.io via (Docker, npm, callin.io cloud, desktop app):
  • Operating system:

Please provide the details so we can assist you further.

 
Posted : 14/11/2024 2:46 pm
aya
 aya
(@aya)
Posts: 7
Active Member
 

Hi,

Welcome to the community

:tada:

Could you share your workflow or parts of it?

Tip for sharing your workflow in the forum

Pasting your callin.io workflow


Ensure to copy your callin.io workflow and paste it in the code block, that is in between the pairs of triple backticks, which also could be achieved by clicking </> (preformatted text) in the editor and pasting in your workflow.

```
<your workflow>
```

Make sure that you’ve removed any sensitive information from your workflow and include dummy data or pinned data as much as you can!


Please provide the rewritten markdown content *it should be in the markdown format.

 
Posted : 21/11/2024 6:48 am
davidpp
(@davidpp)
Posts: 3
Active Member
Topic starter
 

Hi!

I've just updated to the latest version (1.88) and was able to reproduce the issue by creating a temporary flow using Jira as a tool. (Flow below).

CleanShot 2024-11-21 at 10.47.24

 
Posted : 21/11/2024 3:51 pm
davidpp
(@davidpp)
Posts: 3
Active Member
Topic starter
 

I neglected to mention, this is a read operation. However, if I were to perform a "write" operation with Jira, it would create three issues, reusing the same cached payload from the initial tool call.

 
Posted : 21/11/2024 3:52 pm
Mohammed_Rifad
(@mohammed_rifad)
Posts: 1
New Member
 

I'm encountering similar problems with the AI Agent node when using the tool agent. Has anyone managed to resolve this?

Could someone share their solution?

 
Posted : 10/01/2025 1:15 pm
matpb
(@matpb)
Posts: 1
New Member
 

This is similar with extensive tool calling from sub-workflows.

 
Posted : 24/01/2025 3:37 pm
radiaite
(@radiaite)
Posts: 3
New Member
 

I'm encountering the same issue with a sub-workflow.

Has anyone investigated this yet?

 
Posted : 03/03/2025 11:48 am
Niklas_Hatje
(@niklas_hatje)
Posts: 9
Active Member
 

Hello everyone. Could you please share which version you are currently using? I suspect a bug present in a previous version has been resolved. If not, we might need to investigate this further.

 
Posted : 03/03/2025 11:52 am
radiaite
(@radiaite)
Posts: 3
New Member
 

I'm encountering an issue on self-hosted version 1.80.3 and also on cloud version 1.77.4.

 
Posted : 03/03/2025 11:55 am
radiaite
(@radiaite)
Posts: 3
New Member
 

Okay, after updating to version 1.80.5, it's working perfectly now!

:slight_smile:

Thanks for highlighting the version update!

 
Posted : 03/03/2025 1:00 pm
Niklas_Hatje
(@niklas_hatje)
Posts: 9
Active Member
 

Glad to hear that. Thanks for the update.

 
Posted : 10/03/2025 9:26 am
system
(@system)
Posts: 332
Reputable Member
 

This thread was automatically closed 90 days following the last response. New replies are no longer permitted.

 
Posted : 08/06/2025 9:27 am
Share: