rasa-pro 3.12.5__py3-none-any.whl → 3.12.6.dev2__py3-none-any.whl

rasa/dialogue_understanding/generator/llm_based_command_generator.py

@@ -49,6 +49,7 @@ from rasa.shared.utils.llm import (
     llm_factory,
     resolve_model_client_config,
 )
+from rasa.utils.licensing import get_human_readable_licence_owner
 from rasa.utils.log_utils import log_llm
 
 structlogger = structlog.get_logger()
@@ -92,6 +93,8 @@ class LLMBasedCommandGenerator(
         else:
             self.flow_retrieval = None
 
+        self.user_id = get_human_readable_licence_owner()
+
     ### Abstract methods
     @staticmethod
     @abstractmethod
@@ -331,7 +334,9 @@ class LLMBasedCommandGenerator(
 
     @measure_llm_latency
     async def invoke_llm(
-        self, prompt: Union[List[dict], List[str], str]
+        self,
+        prompt: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
     ) -> Optional[LLMResponse]:
         """Use LLM to generate a response.
 
@@ -344,6 +349,7 @@ class LLMBasedCommandGenerator(
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
 
         Returns:
             An LLMResponse object.
@@ -355,7 +361,7 @@ class LLMBasedCommandGenerator(
             self.config.get(LLM_CONFIG_KEY), self.get_default_llm_config()
         )
         try:
-            return await llm.acompletion(prompt)
+            return await llm.acompletion(prompt, metadata)
         except Exception as e:
             # unfortunately, langchain does not wrap LLM exceptions which means
             # we have to catch all exceptions here

rasa/dialogue_understanding/generator/llm_command_generator.py

@@ -55,7 +55,9 @@ class LLMCommandGenerator(SingleStepLLMCommandGenerator):
         )
 
     async def invoke_llm(
-        self, prompt: Union[List[dict], List[str], str]
+        self,
+        prompt: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
     ) -> Optional[LLMResponse]:
         try:
             return await super().invoke_llm(prompt)

rasa/dialogue_understanding/generator/single_step/compact_llm_command_generator.py

@@ -47,6 +47,10 @@ from rasa.shared.constants import (
     AWS_BEDROCK_PROVIDER,
     AZURE_OPENAI_PROVIDER,
     EMBEDDINGS_CONFIG_KEY,
+    LANGFUSE_CUSTOM_METADATA_DICT,
+    LANGFUSE_METADATA_SESSION_ID,
+    LANGFUSE_METADATA_USER_ID,
+    LANGFUSE_TAGS,
     MAX_TOKENS_CONFIG_KEY,
     PROMPT_TEMPLATE_CONFIG_KEY,
     ROUTE_TO_CALM_SLOT,
@@ -362,7 +366,14 @@ class CompactLLMCommandGenerator(LLMBasedCommandGenerator):
             prompt=flow_prompt,
         )
 
-        response = await self.invoke_llm(flow_prompt)
+        metadata = {
+            LANGFUSE_METADATA_USER_ID: self.user_id,
+            LANGFUSE_METADATA_SESSION_ID: tracker.sender_id if tracker else "",
+            LANGFUSE_CUSTOM_METADATA_DICT: {"component": self.__class__.__name__},
+            LANGFUSE_TAGS: [self.__class__.__name__],
+        }
+
+        response = await self.invoke_llm(flow_prompt, metadata)
         llm_response = LLMResponse.ensure_llm_response(response)
         # The check for 'None' maintains compatibility with older versions
         # of LLMCommandGenerator. In previous implementations, 'invoke_llm'

rasa/hooks.py

@@ -1,8 +1,20 @@
 import argparse
 import logging
+import os
 from typing import TYPE_CHECKING, List, Optional, Text, Union
 
+import litellm
 import pluggy
+import structlog
+
+from rasa.shared.providers.constants import (
+    LANGFUSE_CALLBACK_NAME,
+    LANGFUSE_HOST_ENV_VAR,
+    LANGFUSE_PROJECT_ID_ENV_VAR,
+    LANGFUSE_PUBLIC_KEY_ENV_VAR,
+    LANGFUSE_SECRET_KEY_ENV_VAR,
+    RASA_LANGFUSE_INTEGRATION_ENABLED_ENV_VAR,
+)
 
 # IMPORTANT: do not import anything from rasa here - use scoped imports
 #  this avoids circular imports, as the hooks are used in different places
@@ -18,6 +30,7 @@ if TYPE_CHECKING:
 
 hookimpl = pluggy.HookimplMarker("rasa")
 logger = logging.getLogger(__name__)
+structlogger = structlog.get_logger()
 
 
 @hookimpl  # type: ignore[misc]
@@ -57,6 +70,8 @@ def configure_commandline(cmdline_arguments: argparse.Namespace) -> Optional[Tex
         config.configure_tracing(tracer_provider)
         config.configure_metrics(endpoints_file)
 
+    _init_langfuse_integration()
+
     return endpoints_file
 
 
@@ -115,3 +130,43 @@ def after_server_stop() -> None:
 
     if anon_pipeline is not None:
         anon_pipeline.stop()
+
+
+def _is_langfuse_integration_enabled() -> bool:
+    return (
+        os.environ.get(RASA_LANGFUSE_INTEGRATION_ENABLED_ENV_VAR, "false").lower()
+        == "true"
+    )
+
+
+def _init_langfuse_integration() -> None:
+    if not _is_langfuse_integration_enabled():
+        structlogger.info(
+            "hooks._init_langfuse_integration.disabled",
+            event_info="Langfuse integration is disabled.",
+        )
+        return
+
+    if (
+        not os.environ.get(LANGFUSE_HOST_ENV_VAR)
+        or not os.environ.get(LANGFUSE_PROJECT_ID_ENV_VAR)
+        or not os.environ.get(LANGFUSE_PUBLIC_KEY_ENV_VAR)
+        or not os.environ.get(LANGFUSE_SECRET_KEY_ENV_VAR)
+    ):
+        structlogger.warning(
+            "hooks._init_langfuse_integration.missing_langfuse_keys",
+            event_info=(
+                "Langfuse integration is enabled, but some environment variables"
+                "are missing. Please set LANGFUSE_HOST, LANGFUSE_PROJECT_ID, "
+                "LANGFUSE_PUBLIC_KEY and LANGFUSE_SECRET_KEY environment "
+                "variables to use Langfuse integration."
+            ),
+        )
+        return
+
+    litellm.success_callback = [LANGFUSE_CALLBACK_NAME]
+    litellm.failure_callback = [LANGFUSE_CALLBACK_NAME]
+    structlogger.info(
+        "hooks.langfuse_callbacks_initialized",
+        event_info="Langfuse integration initialized.",
+    )

rasa/monkey_patches.py

@@ -0,0 +1,91 @@
+import os
+import traceback
+from typing import Any, Optional
+
+from litellm.secret_managers.main import str_to_bool
+from packaging.version import Version
+
+
+def litellm_langfuse_logger_init_fixed(
+    self: Any,  # we should not import LangfuseLogger class before we patch it
+    langfuse_public_key: Optional[str] = None,
+    langfuse_secret: Optional[str] = None,
+    langfuse_host: str = "https://cloud.langfuse.com",
+    flush_interval: int = 1,
+) -> None:
+    """Monkeypatched version of LangfuseLogger.__init__ from the LiteLLM library.
+
+    This patched version removes a call that fetched the `project_id` from
+    Langfuse Cloud even when it was already set via environment variables.
+    In the original implementation, this call was made *before* initializing
+    the LangfuseClient, which caused the application to freeze for up to 60 seconds.
+
+    By removing this premature call, the monkeypatch avoids the unnecessary network
+    request and prevents the timeout/freeze issue.
+
+    This workaround can be removed once the underlying bug is resolved in LiteLLM:
+    https://github.com/BerriAI/litellm/issues/7732
+    """
+    try:
+        import langfuse
+        from langfuse import Langfuse
+    except Exception as e:
+        raise Exception(
+            f"\033[91mLangfuse not installed, try running 'pip install langfuse' "
+            f"to fix this error: {e}\n{traceback.format_exc()}\033[0m"
+        )
+    # Instance variables
+    self.secret_key = langfuse_secret or os.getenv("LANGFUSE_SECRET_KEY", "")
+    self.public_key = langfuse_public_key or os.getenv("LANGFUSE_PUBLIC_KEY", "")
+
+    self.langfuse_host = langfuse_host or os.getenv(
+        "LANGFUSE_HOST", "https://cloud.langfuse.com"
+    )
+    self.langfuse_host.replace("http://", "https://")
+    if not self.langfuse_host.startswith("https://"):
+        self.langfuse_host = "https://" + self.langfuse_host
+
+    self.langfuse_release = os.getenv("LANGFUSE_RELEASE")
+    self.langfuse_debug = os.getenv("LANGFUSE_DEBUG")
+    self.langfuse_flush_interval = (
+        os.getenv("LANGFUSE_FLUSH_INTERVAL") or flush_interval
+    )
+
+    parameters = {
+        "public_key": self.public_key,
+        "secret_key": self.secret_key,
+        "host": self.langfuse_host,
+        "release": self.langfuse_release,
+        "debug": self.langfuse_debug,
+        "flush_interval": self.langfuse_flush_interval,  # flush interval in seconds
+    }
+
+    if Version(langfuse.version.__version__) >= Version("2.6.0"):
+        parameters["sdk_integration"] = "litellm"
+
+    self.Langfuse = Langfuse(**parameters)
+
+    if os.getenv("UPSTREAM_LANGFUSE_SECRET_KEY") is not None:
+        upstream_langfuse_debug = (
+            str_to_bool(self.upstream_langfuse_debug)
+            if self.upstream_langfuse_debug is not None
+            else None
+        )
+        self.upstream_langfuse_secret_key = os.getenv("UPSTREAM_LANGFUSE_SECRET_KEY")
+        self.upstream_langfuse_public_key = os.getenv("UPSTREAM_LANGFUSE_PUBLIC_KEY")
+        self.upstream_langfuse_host = os.getenv("UPSTREAM_LANGFUSE_HOST")
+        self.upstream_langfuse_release = os.getenv("UPSTREAM_LANGFUSE_RELEASE")
+        self.upstream_langfuse_debug = os.getenv("UPSTREAM_LANGFUSE_DEBUG")
+        self.upstream_langfuse = Langfuse(
+            public_key=self.upstream_langfuse_public_key,
+            secret_key=self.upstream_langfuse_secret_key,
+            host=self.upstream_langfuse_host,
+            release=self.upstream_langfuse_release,
+            debug=(
+                upstream_langfuse_debug
+                if upstream_langfuse_debug is not None
+                else False
+            ),
+        )
+    else:
+        self.upstream_langfuse = None

rasa/shared/constants.py

@@ -338,3 +338,8 @@ ROLE_SYSTEM = "system"
 # Used for key values in ValidateSlotPatternFlowStackFrame
 REFILL_UTTER = "refill_utter"
 REJECTIONS = "rejections"
+
+LANGFUSE_METADATA_USER_ID = "trace_user_id"
+LANGFUSE_METADATA_SESSION_ID = "session_id"
+LANGFUSE_CUSTOM_METADATA_DICT = "trace_metadata"
+LANGFUSE_TAGS = "tags"

rasa/shared/core/slot_mappings.py

@@ -115,6 +115,18 @@ class SlotMapping(BaseModel):
             )
             data_copy[KEY_RUN_ACTION_EVERY_TURN] = deprecated_action
 
+            structlogger.warning(
+                "slot_mapping.deprecated_action_key_replaced_with_run_action_every_turn",
+                slot_name=slot_name,
+                event_info=f"The `{KEY_ACTION}` key in slot mappings "
+                f"has been replaced with "
+                f"the `{KEY_RUN_ACTION_EVERY_TURN}` key. "
+                f"This will result in the custom action "
+                f"being executed at every conversation turn "
+                f"automatically. Remove the key "
+                f"to avoid this behavior.",
+            )
+
         run_action_every_turn = data_copy.pop(KEY_RUN_ACTION_EVERY_TURN, None)
 
         coexistence_system = data_copy.pop(KEY_COEXISTENCE_SYSTEM, None)

rasa/shared/providers/constants.py

@@ -4,3 +4,12 @@ LITE_LLM_API_KEY_FIELD = "api_key"
 LITE_LLM_API_VERSION_FIELD = "api_version"
 LITE_LLM_MODEL_FIELD = "model"
 LITE_LLM_AZURE_AD_TOKEN = "azure_ad_token"
+
+# Enable or disable Langfuse integration
+RASA_LANGFUSE_INTEGRATION_ENABLED_ENV_VAR = "RASA_LANGFUSE_INTEGRATION_ENABLED"
+# Langfuse configuration
+LANGFUSE_CALLBACK_NAME = "langfuse"
+LANGFUSE_HOST_ENV_VAR = "LANGFUSE_HOST"
+LANGFUSE_PROJECT_ID_ENV_VAR = "LANGFUSE_PROJECT_ID"
+LANGFUSE_PUBLIC_KEY_ENV_VAR = "LANGFUSE_PUBLIC_KEY"
+LANGFUSE_SECRET_KEY_ENV_VAR = "LANGFUSE_SECRET_KEY"

rasa/shared/providers/llm/_base_litellm_client.py

@@ -2,7 +2,7 @@ from __future__ import annotations
 
 import logging
 from abc import abstractmethod
-from typing import Any, Dict, List, Union, cast
+from typing import Any, Dict, List, Optional, Union, cast
 
 import structlog
 from litellm import acompletion, completion, validate_environment
@@ -120,7 +120,11 @@ class _BaseLiteLLMClient:
             raise ProviderClientValidationError(event_info)
 
     @suppress_logs(log_level=logging.WARNING)
-    def completion(self, messages: Union[List[dict], List[str], str]) -> LLMResponse:
+    def completion(
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
+    ) -> LLMResponse:
         """Synchronously generate completions for given list of messages.
 
         Args:
@@ -132,6 +136,7 @@ class _BaseLiteLLMClient:
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
 
         Returns:
             List of message completions.
@@ -149,7 +154,9 @@ class _BaseLiteLLMClient:
 
     @suppress_logs(log_level=logging.WARNING)
     async def acompletion(
-        self, messages: Union[List[dict], List[str], str]
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
     ) -> LLMResponse:
         """Asynchronously generate completions for given list of messages.
 
@@ -162,6 +169,7 @@ class _BaseLiteLLMClient:
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
 
         Returns:
             List of message completions.
@@ -172,7 +180,9 @@ class _BaseLiteLLMClient:
         try:
             formatted_messages = self._get_formatted_messages(messages)
             arguments = resolve_environment_variables(self._completion_fn_args)
-            response = await acompletion(messages=formatted_messages, **arguments)
+            response = await acompletion(
+                messages=formatted_messages, metadata=metadata, **arguments
+            )
             return self._format_response(response)
         except Exception as e:
             message = ""

rasa/shared/providers/llm/litellm_router_llm_client.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 import logging
-from typing import Any, Dict, List, Union
+from typing import Any, Dict, List, Optional, Union
 
 import structlog
 
@@ -122,9 +122,12 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
             raise ProviderClientAPIException(e)
 
     @suppress_logs(log_level=logging.WARNING)
-    def completion(self, messages: Union[List[dict], List[str], str]) -> LLMResponse:
-        """
-        Synchronously generate completions for given list of messages.
+    def completion(
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
+    ) -> LLMResponse:
+        """Synchronously generate completions for given list of messages.
 
         Method overrides the base class method to call the appropriate
         completion method based on the configuration. If the chat completions
@@ -140,8 +143,11 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """
@@ -158,10 +164,11 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
 
     @suppress_logs(log_level=logging.WARNING)
     async def acompletion(
-        self, messages: Union[List[dict], List[str], str]
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
     ) -> LLMResponse:
-        """
-        Asynchronously generate completions for given list of messages.
+        """Asynchronously generate completions for given list of messages.
 
         Method overrides the base class method to call the appropriate
         completion method based on the configuration. If the chat completions
@@ -177,8 +184,11 @@ class LiteLLMRouterLLMClient(_BaseLiteLLMRouterClient, _BaseLiteLLMClient):
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
+
         Returns:
             List of message completions.
+
         Raises:
             ProviderClientAPIException: If the API request fails.
         """

rasa/shared/providers/llm/llm_client.py

@@ -1,21 +1,19 @@
 from __future__ import annotations
 
-from typing import Dict, List, Protocol, Union, runtime_checkable
+from typing import Any, Dict, List, Optional, Protocol, Union, runtime_checkable
 
 from rasa.shared.providers.llm.llm_response import LLMResponse
 
 
 @runtime_checkable
 class LLMClient(Protocol):
-    """
-    Protocol for an LLM client that specifies the interface for interacting
+    """Protocol for an LLM client that specifies the interface for interacting
     with the API.
     """
 
     @classmethod
     def from_config(cls, config: dict) -> LLMClient:
-        """
-        Initializes the llm client with the given configuration.
+        """Initializes the llm client with the given configuration.
 
         This class method should be implemented to parse the given
         configuration and create an instance of an llm client.
@@ -24,17 +22,24 @@ class LLMClient(Protocol):
 
     @property
     def config(self) -> Dict:
-        """
-        Returns the configuration for that the llm client is initialized with.
+        """Returns the configuration for that the llm client is initialized with.
 
         This property should be implemented to return a dictionary containing
         the configuration settings for the llm client.
         """
         ...
 
-    def completion(self, messages: Union[List[dict], List[str], str]) -> LLMResponse:
-        """
-        Synchronously generate completions for given list of messages.
+    def completion(
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
+    ) -> LLMResponse:
+        """Synchronously generate completions for given list of messages.
+        def completion(
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
+        ) -> LLMResponse:
 
         This method should be implemented to take a list of messages (as
         strings) and return a list of completions (as strings).
@@ -48,16 +53,19 @@ class LLMClient(Protocol):
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
+
         Returns:
             LLMResponse
         """
         ...
 
     async def acompletion(
-        self, messages: Union[List[dict], List[str], str]
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
     ) -> LLMResponse:
-        """
-        Asynchronously generate completions for given list of messages.
+        """Asynchronously generate completions for given list of messages.
 
         This method should be implemented to take a list of messages (as
         strings) and return a list of completions (as strings).
@@ -71,14 +79,15 @@ class LLMClient(Protocol):
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
+
         Returns:
             LLMResponse
         """
         ...
 
     def validate_client_setup(self, *args, **kwargs) -> None:  # type: ignore
-        """
-        Perform client setup validation.
+        """Perform client setup validation.
 
         This method should be implemented to validate whether the client can be
         used with the parameters provided through configuration or environment

rasa/shared/providers/llm/self_hosted_llm_client.py

@@ -237,7 +237,9 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
             raise ProviderClientAPIException(e)
 
     async def acompletion(
-        self, messages: Union[List[dict], List[str], str]
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
     ) -> LLMResponse:
         """Asynchronous completion of the model with the given messages.
 
@@ -255,6 +257,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
                 - a list of messages. Each message is a string and will be formatted
                     as a user message.
                 - a single message as a string which will be formatted as user message.
+            metadata: Optional metadata to be passed to the LLM call.
 
         Returns:
             The completion response.
@@ -263,7 +266,11 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
             return await super().acompletion(messages)
         return await self._atext_completion(messages)
 
-    def completion(self, messages: Union[List[dict], List[str], str]) -> LLMResponse:
+    def completion(
+        self,
+        messages: Union[List[dict], List[str], str],
+        metadata: Optional[Dict[str, Any]] = None,
+    ) -> LLMResponse:
         """Completion of the model with the given messages.
 
         Method overrides the base class method to call the appropriate
@@ -273,6 +280,7 @@ class SelfHostedLLMClient(_BaseLiteLLMClient):
 
         Args:
             messages: The messages to be used for completion.
+            metadata: Optional metadata to be passed to the LLM call.
 
         Returns:
             The completion response.

rasa/shared/utils/health_check/health_check.py

@@ -3,6 +3,7 @@ import sys
 from typing import Any, Dict, Optional
 
 from rasa.shared.constants import (
+    LANGFUSE_CUSTOM_METADATA_DICT,
     LLM_API_HEALTH_CHECK_DEFAULT_VALUE,
     LLM_API_HEALTH_CHECK_ENV_VAR,
     MODELS_CONFIG_KEY,
@@ -198,7 +199,12 @@ def send_test_llm_api_request(
         config=llm_client.config,
     )
     try:
-        llm_client.completion("hello")
+        llm_client.completion(
+            "hello",
+            metadata={
+                LANGFUSE_CUSTOM_METADATA_DICT: {"component": log_source_component}
+            },
+        )
     except Exception as e:
         structlogger.error(
             f"{log_source_function}.send_test_llm_api_request_failed",

rasa/tracing/instrumentation/attribute_extractors.py

@@ -372,6 +372,7 @@ def extract_llm_config(
 def extract_attrs_for_llm_based_command_generator(
     self: "LLMBasedCommandGenerator",
     prompt: str,
+    metadata: Optional[Dict[str, Any]] = None,
 ) -> Dict[str, Any]:
     from rasa.dialogue_understanding.generator.flow_retrieval import (
         DEFAULT_EMBEDDINGS_CONFIG,
@@ -387,8 +388,7 @@ def extract_attrs_for_llm_based_command_generator(
 
 
 def extract_attrs_for_contextual_response_rephraser(
-    self: Any,
-    prompt: str,
+    self: Any, prompt: str, sender_id: str
 ) -> Dict[str, Any]:
     from rasa.core.nlg.contextual_response_rephraser import DEFAULT_LLM_CONFIG
 
@@ -721,7 +721,7 @@ def extract_attrs_for_intentless_policy_find_closest_response(
 
 
 def extract_attrs_for_intentless_policy_generate_llm_answer(
-    self: "IntentlessPolicy", llm: "BaseLLM", prompt: str
+    self: "IntentlessPolicy", llm: "BaseLLM", prompt: str, sender_id: str
 ) -> Dict[str, Any]:
     from rasa.core.policies.intentless_policy import (
         DEFAULT_EMBEDDINGS_CONFIG,
@@ -738,7 +738,7 @@ def extract_attrs_for_intentless_policy_generate_llm_answer(
 
 
 def extract_attrs_for_enterprise_search_generate_llm_answer(
-    self: "EnterpriseSearchPolicy", llm: "BaseLLM", prompt: str
+    self: "EnterpriseSearchPolicy", llm: "BaseLLM", prompt: str, sender_id: str
 ) -> Dict[str, Any]:
     from rasa.core.policies.enterprise_search_policy import (
         DEFAULT_EMBEDDINGS_CONFIG,

rasa/tracing/instrumentation/intentless_policy_instrumentation.py

@@ -121,12 +121,13 @@ def _instrument_generate_answer(
             response_examples: List[str],
             conversation_samples: List[str],
             history: str,
+            sender_id: str,
         ) -> Optional[str]:
             with tracer.start_as_current_span(
                 f"{self.__class__.__name__}.{fn.__name__}"
             ) as span:
                 llm_response = await fn(
-                    self, response_examples, conversation_samples, history
+                    self, response_examples, conversation_samples, history, sender_id
                 )
                 span.set_attributes(
                     {

rasa/utils/licensing.py

@@ -539,3 +539,18 @@ async def _count_conversations_after(
         return 0
 
     return await tracker_store.count_conversations(after_timestamp=after_timestamp)
+
+
+def get_human_readable_licence_owner() -> str:
+    user_id = "unknown"
+
+    try:
+        retrieved_license = retrieve_license_from_env()
+        if retrieved_license:
+            decoded = License.decode(retrieved_license)
+            if decoded:
+                user_id = (
+                    f"{decoded.company or ''}_{decoded.email or ''}_{decoded.jti or ''}"
+                )
+    finally:
+        return user_id

rasa/version.py

@@ -1,3 +1,3 @@
 # this file will automatically be changed,
 # do not add anything but the version number here!
-__version__ = "3.12.5"
+__version__ = "3.12.6.dev2"