feat: OTel aligned attributes checks

[constantinius]

Closes https://linear.app/getsentry/issue/TET-1763/add-new-otel-attributes-to-the-testing-framework

[linear]

TET-1763 Add new OTel attributes to the testing framework

[github-actions]

🔴 AI SDK Integration Test Results

Status: 2 regressions detected

Summary

Metric	main	PR	Change
Total Tests	159	159	—
Passed	50	48	-2 ⚠️
Failed	109	111	+2 ⚠️

🔴 Regressions

These tests were passing on main but are now failing:

js/mastra :: Tool Call Agent Test

Error: 2 check(s) failed:

2 check(s) failed:
expected { result: 8 } to deeply equal 8
Should have at least 2 tool call(s) in output messages: expected +0 to be at least 2

js/mastra :: Tool Error Agent Test

Error: 1 check(s) failed:

1 check(s) failed:
Should have at least 1 tool call(s) in output messages: expected +0 to be at least 1

Test Matrix

SDK	Basic Agent Test	Basic Agent Test (async)	Basic Agent Test (sync)	Basic Error LLM Test (async, blocking)	Basic Error LLM Test (async, streaming)	Basic Error LLM Test (blocking)	Basic Error LLM Test (streaming)	Basic Error LLM Test (sync, blocking)	Basic Error LLM Test (sync, streaming)	Basic LLM Test (async, blocking)	Basic LLM Test (async, streaming)	Basic LLM Test (blocking)	Basic LLM Test (streaming)	Basic LLM Test (sync, blocking)	Basic LLM Test (sync, streaming)	Long Input Agent Test	Long Input Agent Test (async)	Long Input Agent Test (sync)	Long Input LLM Test (async, blocking)	Long Input LLM Test (async, streaming)	Long Input LLM Test (blocking)	Long Input LLM Test (streaming)	Long Input LLM Test (sync, blocking)	Long Input LLM Test (sync, streaming)	Multi-Turn LLM Test (async, blocking)	Multi-Turn LLM Test (async, streaming)	Multi-Turn LLM Test (blocking)	Multi-Turn LLM Test (streaming)	Multi-Turn LLM Test (sync, blocking)	Multi-Turn LLM Test (sync, streaming)	Tool Call Agent Test	Tool Call Agent Test (async)	Tool Call Agent Test (sync)	Tool Error Agent Test	Tool Error Agent Test (async)	Tool Error Agent Test (sync)	Vision Agent Test	Vision Agent Test (async)	Vision Agent Test (sync)	Vision LLM Test (async, blocking)	Vision LLM Test (async, streaming)	Vision LLM Test (blocking)	Vision LLM Test (streaming)	Vision LLM Test (sync, blocking)	Vision LLM Test (sync, streaming)
js/anthropic	—	—	—	—	—	✅	✅	—	—	—	—	❌	❌	—	—	—	—	—	—	—	❌	❌	—	—	—	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	—
js/google-genai	—	—	—	—	—	❌	❌	—	—	—	—	❌	❌	—	—	—	—	—	—	—	❌	❌	—	—	—	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	—
js/langchain	—	—	—	—	—	✅	✅	—	—	—	—	❌	❌	—	—	—	—	—	—	—	❌	❌	—	—	—	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	—
js/mastra	✅	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	❌📉	—	—	❌📉	—	—	✅	—	—	—	—	—	—	—	—
js/openai	—	—	—	—	—	✅	✅	—	—	—	—	✅	❌	—	—	—	—	—	—	—	❌	❌	—	—	—	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	—
js/vercel	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	—	—	❌	—	—	❌	—	—	—	—	—	—	—	—
py/anthropic	—	—	—	✅	❌	—	—	✅	❌	✅	❌	—	—	✅	❌	—	—	—	❌	❌	—	—	❌	❌	✅	❌	—	—	✅	❌	—	—	—	—	—	—	—	—	—	✅	❌	—	—	✅	❌
py/google-genai	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	❌	❌	—	❌	❌	—	—	—	—	—	—
py/langchain	—	—	—	✅	✅	—	—	✅	✅	✅	✅	—	—	✅	✅	—	—	—	❌	❌	—	—	❌	❌	✅	✅	—	—	✅	✅	—	—	—	—	—	—	—	—	—	✅	✅	—	—	✅	✅
py/langgraph	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	❌	—	❌	❌	—	❌	❌	—	—	—	—	—	—
py/litellm	—	—	—	❌	❌	—	—	❌	❌	❌	❌	—	—	✅	✅	—	—	—	❌	❌	—	—	❌	❌	❌	❌	—	—	✅	✅	—	—	—	—	—	—	—	—	—	❌	❌	—	—	✅	✅
py/openai	—	—	—	✅	✅	—	—	✅	✅	✅	❌	—	—	✅	❌	—	—	—	❌	❌	—	—	❌	❌	✅	❌	—	—	✅	❌	—	—	—	—	—	—	—	—	—	❌	❌	—	—	❌	❌
py/openai-agents	—	✅	—	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	—	—	❌	—	—	❌	—	—	—	—	—	—	—
py/pydantic-ai	—	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	—	—	—	—	—	—	—	—	—	—	—	—	—	—	❌	—	—	❌	—	—	❌	—	—	—	—	—	—	—

Legend: ✅ Pass | ❌ Fail | ✅🔧 Fixed | ❌📉 Regressed | ✅🆕 New (pass) | ❌🆕 New (fail) | 🗑️ Removed

---

Generated by AI SDK Integration Tests