You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
it("includes gen_ai.request.model, gen_ai.provider.name, gen_ai.operation.name and gen_ai.workflow.name on the agent span from aw_info.json",async()=>{
1890
+
it("includes gen_ai.request.model, gen_ai.system, gh-aw.engine, gen_ai.operation.name and gen_ai.workflow.name on the agent span from aw_info.json",async()=>{
it("omits gen_ai.request.model, gen_ai.provider.name and gen_ai.workflow.name from the agent span when model, engine_id and workflow_name are absent",async()=>{
1923
+
it("omits gen_ai.request.model, gen_ai.system, gh-aw.engine and gen_ai.workflow.name from the agent span when model, engine_id and workflow_name are absent",async()=>{
> Prior to v0.70, the agent span used private `gh-aw.*` attribute names (`gh-aw.model`, `gh-aw.tokens.input`, etc.) and `SPAN_KIND_INTERNAL`. These attributes were removed and replaced with the `gen_ai.*` convention above. Update any dashboards or alert rules that reference the old attribute names.
899
900
901
+
> [!NOTE]
902
+
> Prior to v0.76, the engine was emitted as `gen_ai.provider.name` with the raw gh-aw engine ID. It is now emitted as the standard `gen_ai.system` attribute with a mapped OTel system name, and the raw engine ID is preserved in `gh-aw.engine`.
0 commit comments