GenAI Utils | Add support for ToolCall Invocations

[keith-decker]

Description

Implementing the ToolCall handler methods. Check out the example app here

Some output from the example app:

Trace ID: 9db02aa0f3f04e3280383277aeb8daa1
└── Span ID: a6aac8039683ad5f (Parent: none) - Name: agent_workflow [op:invoke_agent] (Type: span)
    ├── Span ID: f225d302435ccab6 (Parent: a6aac8039683ad5f) - Name: chat gpt-4o-mini [op:chat] (Type: span)
    │   ├── Metric: gen_ai.client.operation.duration [op:chat] (Type: metric)
    │   ├── Metric: gen_ai.client.token.usage (input) [op:chat] (Type: metric)
    │   └── Metric: gen_ai.client.token.usage (output) [op:chat] (Type: metric)
    └── Span ID: a56510b4b1a902c5 (Parent: a6aac8039683ad5f) - Name: execute_tool get_weather [op:execute_tool] (Type: span)

Span #1
    Trace ID       : 9db02aa0f3f04e3280383277aeb8daa1
    Parent ID      : a6aac8039683ad5f
    ID             : a56510b4b1a902c5
    Name           : execute_tool get_weather
    Kind           : Internal
    Start time     : 2026-03-24 21:12:59.832384 +0000 UTC
    End time       : 2026-03-24 21:12:59.833126 +0000 UTC
    Status code    : Ok
    Status message : 
Attributes:
     -> gen_ai.operation.name: Str(execute_tool)
     -> gen_ai.tool.name: Str(get_weather)
     -> gen_ai.tool.call.id: Str(019d21b1-96b8-7f00-961c-e2e9c1a0baa1)
     -> gen_ai.tool.type: Str(function)
     -> gen_ai.tool.description: Str(Get current weather for a location.

    Args:
        location: The city name to get weather for.

    Returns:
        Weather information as a string.)

Fixes # (issue)

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• Test A

Does This PR Require a Core Repo Change?

• Yes. - Link to PR:
• No.

Checklist:

See contributing.md for styleguide, changelog guidelines, and more.

• Followed the style guidelines of this project
• Changelogs have been updated
• Unit tests have been added
• Documentation has been updated

[DylanRussell]

In the google gen AI instrumentation we monkey patch the actual function call: https://github.com/open-telemetry/opentelemetry-python-contrib/blob/main/instrumentation-genai/opentelemetry-instrumentation-google-genai/src/opentelemetry/instrumentation/google_genai/tool_call_wrapper.py#L164

IDK if it makes sense to do something like that here or not (accept the actual Callable object)?

Either way I'll try to update that instrumentation to use this once we submit

[eternalcuriouslearner]

Overall LGTM!! I left few questions to understand the design better.

[eternalcuriouslearner]

question: Why are we ignoring attributes: dict[str, Any] = field(default_factory=_new_str_any_dict) attribute here?

[keith-decker]

These are properties used by the metrics. the attributes you mentioned is used for custom attributes the user might want to use on the span.

[eternalcuriouslearner]

Ahh dumb question are we ignoring custom attributes for ToolCall?