Skip to content

Remove inference attributes from agent invocation span #2632

@aabmass

Description

@aabmass

Area(s)

area:gen-ai

What's missing?

N/A

Describe the solution you'd like

This came up in #2179 (comment).

span.gen_ai.invoke_agent.client includes inference attributes by extending attributes.gen_ai.inference.client

- id: span.gen_ai.invoke_agent.client
type: span
stability: development
span_kind: client
brief: >
Describes GenAI agent invocation and is usually applicable when working
with remote agent services.
note: |
The `gen_ai.operation.name` SHOULD be `invoke_agent`.
The **span name** SHOULD be `invoke_agent {gen_ai.agent.name}` if `gen_ai.agent.name` is readily available.
When `gen_ai.agent.name` is not available, it SHOULD be `invoke_agent`.
Semantic conventions for individual GenAI systems and frameworks MAY specify different span name format.
extends: attributes.gen_ai.inference.client

This has been there since the span was introduced in #1900. Agent invocations don't really have the same input/output as a raw inference call and are semantically distinct IMO. It's creating some mismatch in the expected formats. I propose removing the extends altogether.

Tip

React with 👍 to help prioritize this issue. Please use comments to provide useful context, avoiding +1 or me too, to help us triage it. Learn more here.

Metadata

Metadata

Assignees

Type

No type

Projects

Status

Need triage

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions