feat(contrib/mcp-go): Trace MCP session initializations with MLObs spans

[jboolean]

NB: This is part of a stack

What does this PR do?

In addition to tool calls, record task spans for session initialization.

Motivation

We want to record data like clientInfo during session Initialization and this seems like the best way to do it.

I considered a metric, but in the future we will also want to tag the session ID, and metrics do not do well with unique tags. Also, this allows viewing everything together in the LLMObs UI.

A writeup with more context on this design

Relates to MLOB-4373

Future

In future PRs I will add tags for client info and session ID.

Reviewer's Checklist

• Changed code has unit tests for its functionality at or near 100% coverage.
• System-Tests covering this feature have been added and enabled with the va.b.c-dev version tag.
• There is a benchmark for any new code, or changes to existing code.
• If this interacts with the agent in a new way, a system test has been added.
• New code is free of linting errors. You can check this by running ./scripts/lint.sh locally.
• Add an appropriate team label so this PR gets put in the right place for the release notes.
• Non-trivial go.mod changes, e.g. adding new modules, are reviewed by @DataDog/dd-trace-go-guild.

Unsure? Have a question? Request a review!

[jboolean]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• feat(contrib/mcp-go): Record MCP Session ID as tag #4106
• feat(contrib/mcp-go): Trace MCP session initializations with MLObs spans #4101 👈 (View in Graphite)
• feat(contrib/mcp-go): Initial mcp-go tracer implementation #4100
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[pr-commenter]

Benchmarks

Benchmark execution time: 2025-11-07 19:57:54

Comparing candidate commit 5f20214 in PR branch jb/contrib-mcp-go-init-span with baseline commit 65cc8e1 in branch jb/contrib-mcp-go-base.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 24 metrics, 0 unstable metrics.

[rarguelloF]

is this expected to happen often? I think at least we should log something (debug if this is expected to happen somewhat frequently, or warn if not).

also, should we annotate the input with the error or some special message to let customers know something went wrong directly while viewing the span?

[rarguelloF]

same here, i think we should handle these errors somehow

[rarguelloF]

should we also finish spans that were evicted for other reasons?

[datadog-official]

⚠️ Tests

⚠️ Warnings

❄️ 1 New flaky test detected

TestWrapConsumerGroupHandler from github.com/DataDog/dd-trace-go/contrib/IBM/sarama/v2 (Datadog)

Failed

=== RUN   TestWrapConsumerGroupHandler
2025/11/07 19:51:11 Sarama consumer up and running!...
    consumer_group_test.go:146: Message claimed: value = test 1, timestamp = 2025-11-07 19:51:11.207 +0000 UTC, topic = IBM_sarama_TestWrapConsumerGroupHandler
    consumer_group_test.go:88: 
        	Error Trace:	/home/runner/work/dd-trace-go/dd-trace-go/contrib/IBM/sarama/consumer_group_test.go:88
        	Error:      	"[
        	            	name: kafka.produce
        	            	tags: map[string]interface {}{"_dd.base_service":"", "_dd.p.tid":"690e4daf00000000", "_dd.profiling.enabled":0, "_dd.top_level":1, "component":"IBM/sarama", "language":"go", "messaging.destination.name":"IBM_sarama_TestWrapConsumerGroupHandler", "messaging.kafka.partition":0, "messaging.system":"kafka", "offset":0, "resource.name":"Produce Topic IBM_sarama_TestWrapConsumerGroupHandler", "service.name":"kafka", "span.kind":"producer", "span.name":"kafka.produce", "span.type":"queue"}
...

ℹ️ Info

🧪 All tests passed

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 5f20214 | Docs | Datadog PR Page | Was this helpful? Give us feedback!}