Skip to content

Add thinking messages to Sentry.Extensions.AI #5267

@jankalfus

Description

@jankalfus

Problem Statement

Hi, we have instrumentation for Anthropic.SDK configured according to https://docs.sentry.io/platforms/dotnet/tracing/instrumentation/ai-agents-module/

It works nicely! One thing we’re missing currently though, are the thinking messages. Only the assistant messages seem to get captured in the output. Without the thinking messages, when an LLM fails to do something because it took a wrong approach, we don’t know why it took the route it took. So having the thinking messages captured and displayed in the UI would be really useful in such cases.

Solution Brainstorm

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    No fields configured for issues without a type.

    Projects

    Status

    Waiting for: Product Owner

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions