You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@camel.apache.org by "Claus Ibsen (Jira)" <ji...@apache.org> on 2026/09/22 06:35:00 UTC

[jira] [Commented] (CAMEL-24721) camel-openai: token budget and agentic observability for the Responses tool loop

    [ https://issues.apache.org/jira/browse/CAMEL-24721?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18117912#comment-18117912 ] 

Claus Ibsen commented on CAMEL-24721:
-------------------------------------

Merged in https://github.com/apache/camel/pull/26707 (commit 362d112c7cbb), fix version 4.23.0.

_Claude Code on behalf of davsclaus_

> camel-openai: token budget and agentic observability for the Responses tool loop
> --------------------------------------------------------------------------------
>
>                 Key: CAMEL-24721
>                 URL: https://issues.apache.org/jira/browse/CAMEL-24721
>             Project: Camel
>          Issue Type: Improvement
>          Components: camel-openai
>            Reporter: Federico Mariani
>            Assignee: Karol Krawczyk
>            Priority: Major
>             Fix For: 4.23.0
>
>
> CAMEL-24670 adds a tool loop to the {{responses}} operation, but only {{chat-completion}} supports the agentic extras:
> * The cumulative {{CamelOpenAIAgenticPromptTokens}}, {{CamelOpenAIAgenticCompletionTokens}} and {{CamelOpenAIAgenticTotalTokens}} headers are not set, so after a tool loop the token headers only reflect the last model call.
> * {{maxAgenticTokens}} is not enforced.
> * The agentic trace ({{CamelOpenAIAgenticTrace}}) and the agentic loop lifecycle events are not produced.
> Support them in the Responses tool loop as {{chat-completion}} does.
> _Claude Code on behalf of Federico Mariani (Croway)_



--
This message was sent by Atlassian Jira
(v8.20.10#820010)