Refactor: use unified utility function for $ai_lib_metadata

Creates a single `get_ai_lib_metadata(framework)` utility function to generate
the $ai_lib_metadata object, replacing inline implementations across the
codebase.

Changes:
- Add get_ai_lib_metadata() utility to utils.py
- Update LangChain callbacks to use utility function
- Update call_llm_and_track_usage() to use utility function
- Update call_llm_and_track_usage_async() to use utility function
- Update capture_streaming_event() to use utility function

Benefits:
- Consistency across all integrations
- Single source of truth for metadata structure
- Easier to extend with version detection later

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <[email protected]>

Author: andrewm4894

posthog/ai/langchain/callbacks.py

@@ -36,7 +36,7 @@
 from pydantic import BaseModel
 
 from posthog import setup
-from posthog.ai.utils import get_model_params, with_privacy_mode
+from posthog.ai.utils import get_model_params, with_privacy_mode, get_ai_lib_metadata
 from posthog.ai.sanitization import sanitize_langchain
 from posthog.client import Client
 
@@ -486,7 +486,7 @@ def _capture_trace_or_span(
             "$ai_latency": run.latency,
             "$ai_span_name": run.name,
             "$ai_span_id": run_id,
-            "$ai_lib_metadata": _get_ai_lib_metadata(),
+            "$ai_lib_metadata": get_ai_lib_metadata("langchain"),
         }
         if parent_run_id is not None:
             event_properties["$ai_parent_id"] = parent_run_id
@@ -557,7 +557,7 @@ def _capture_generation(
             "$ai_http_status": 200,
             "$ai_latency": run.latency,
             "$ai_base_url": run.base_url,
-            "$ai_lib_metadata": _get_ai_lib_metadata(),
+            "$ai_lib_metadata": get_ai_lib_metadata("langchain"),
         }
 
         if run.tools:
@@ -867,12 +867,3 @@ def _stringify_exception(exception: BaseException) -> str:
     if description:
         return f"{exception.__class__.__name__}: {description}"
     return exception.__class__.__name__
-
-
-def _get_ai_lib_metadata() -> Dict[str, Any]:
-    """
-    Generate the $ai_lib_metadata dict with framework information.
-    Since this file is langchain/callbacks.py, we know the framework is LangChain.
-    Returns a dict with schema version and framework name.
-    """
-    return {"schema": "v1", "frameworks": [{"name": "langchain"}]}

posthog/ai/utils.py

@@ -12,6 +12,14 @@
 )
 
 
+def get_ai_lib_metadata(framework: str) -> Dict[str, Any]:
+    """
+    Generate AI library metadata object with framework name.
+    Used to identify which framework/SDK generated the AI event.
+    """
+    return {"schema": "v1", "frameworks": [{"name": framework}]}
+
+
 def merge_usage_stats(
     target: TokenUsage, source: TokenUsage, mode: str = "incremental"
 ) -> None:
@@ -320,7 +328,7 @@ def call_llm_and_track_usage(
                 ph_client, posthog_privacy_mode, kwargs.get("instructions")
             )
 
-        event_properties["$ai_lib_metadata"] = {"schema": "v1", "frameworks": [{"name": provider}]}
+        event_properties["$ai_lib_metadata"] = get_ai_lib_metadata(provider)
 
         # send the event to posthog
         if hasattr(ph_client, "capture") and callable(ph_client.capture):
@@ -425,7 +433,7 @@ async def call_llm_and_track_usage_async(
                 ph_client, posthog_privacy_mode, kwargs.get("instructions")
             )
 
-        event_properties["$ai_lib_metadata"] = {"schema": "v1", "frameworks": [{"name": provider}]}
+        event_properties["$ai_lib_metadata"] = get_ai_lib_metadata(provider)
 
         # send the event to posthog
         if hasattr(ph_client, "capture") and callable(ph_client.capture):
@@ -487,7 +495,7 @@ def capture_streaming_event(
 
     # Build base event properties
     event_properties = {
-        "$ai_lib_metadata": {"schema": "v1", "frameworks": [{"name": event_data["provider"]}]},
+        "$ai_lib_metadata": get_ai_lib_metadata(event_data["provider"]),
         "$ai_provider": event_data["provider"],
         "$ai_model": event_data["model"],
         "$ai_model_parameters": get_model_params(event_data["kwargs"]),