ibm-watsonx-orchestrate 1.9.0b2__py3-none-any.whl → 1.10.0b1__py3-none-any.whl

ibm_watsonx_orchestrate/cli/main.py

@@ -16,6 +16,7 @@ from ibm_watsonx_orchestrate.cli.commands.knowledge_bases.knowledge_bases_comman
 from ibm_watsonx_orchestrate.cli.commands.toolkit.toolkit_command import toolkits_app
 from ibm_watsonx_orchestrate.cli.commands.evaluations.evaluations_command import evaluation_app
 from ibm_watsonx_orchestrate.cli.commands.copilot.copilot_command import copilot_app
+from ibm_watsonx_orchestrate.cli.commands.voice_configurations.voice_configurations_command import voice_configurations_app
 from ibm_watsonx_orchestrate.cli.init_helper import init_callback
 
 import urllib3
@@ -35,6 +36,7 @@ app.add_typer(tools_app, name="tools", help='Interact with the tools in your act
 app.add_typer(toolkits_app, name="toolkits", help="Interact with the toolkits in your active env")
 app.add_typer(knowledge_bases_app, name="knowledge-bases", help="Upload knowledge your agents can search through to your active env")
 app.add_typer(connections_app, name="connections", help='Interact with the agents in your active env')
+app.add_typer(voice_configurations_app, name="voice-configs", help="Configure voice providers to enable voice interaction with your agents")
 app.add_typer(server_app, name="server", help='Manipulate your local Orchestrate Developer Edition server [requires entitlement]')
 app.add_typer(chat_app, name="chat", help='Launch the chat ui for your local Developer Edition server [requires entitlement]')
 app.add_typer(models_app, name="models", help='List the available large language models (llms) that can be used in your agent definitions')

ibm_watsonx_orchestrate/client/agents/agent_client.py

@@ -1,7 +1,26 @@
 from ibm_watsonx_orchestrate.client.base_api_client import BaseAPIClient, ClientAPIException
 from typing_extensions import List, Optional
+from enum import Enum
+
 from ibm_watsonx_orchestrate.client.utils import is_local_dev
 from pydantic import BaseModel
+import time
+import logging
+
+logger = logging.getLogger(__name__)
+
+POLL_INTERVAL = 2
+MAX_RETRIES = 10
+
+class ReleaseMode(str, Enum):
+    DEPLOY = "deploy"
+    UNDEPLOY = "undeploy"
+
+class ReleaseStatus(str, Enum):
+    SUCCESS = "success"
+    NONE = "none"
+    FAILED = "failed"
+    IN_PROGRESS = "in_progress"
 
 def transform_agents_from_flat_agent_spec(agents: dict | list[dict] ) -> dict | list[dict]:
     if isinstance(agents,list):
@@ -83,7 +102,6 @@ class AgentClient(BaseAPIClient):
         super().__init__(*args, **kwargs)
         self.base_endpoint = "/orchestrate/agents" if is_local_dev(self.base_url) else "/agents"
 
-
     def create(self, payload: dict) -> AgentUpsertResponse:
         response = self._post(self.base_endpoint, data=transform_agents_from_flat_agent_spec(payload))
         return AgentUpsertResponse.model_validate(response)
@@ -120,4 +138,49 @@ class AgentClient(BaseAPIClient):
     def get_drafts_by_ids(self, agent_ids: List[str]) -> List[dict]:
         formatted_agent_ids = [f"ids={x}" for x  in agent_ids]
         return transform_agents_to_flat_agent_spec(self._get(f"{self.base_endpoint}?{'&'.join(formatted_agent_ids)}&include_hidden=true"))
+
+    def poll_release_status(self, agent_id: str, environment_id: str, mode: str = "deploy") -> bool:
+        expected_status = {
+            ReleaseMode.DEPLOY: ReleaseStatus.SUCCESS,
+            ReleaseMode.UNDEPLOY: ReleaseStatus.NONE
+        }[mode]
+
+        for attempt in range(MAX_RETRIES):
+            try:
+                response = self._get(
+                    f"{self.base_endpoint}/{agent_id}/releases/status?environment_id={environment_id}"
+                )
+            except Exception as e:
+                logger.error(f"Polling for Deployment/Undeployment failed on attempt {attempt + 1}: {e}")
+                return False
+
+            if not isinstance(response, dict):
+                logger.warning(f"Invalid response format: {response}")
+                return False
+            
+            status = response.get("deployment_status")
+
+            if status == expected_status:
+                return True
+            elif status == "failed":
+                return False
+            elif status == "in_progress":
+                pass
+
+            time.sleep(POLL_INTERVAL)
+
+        logger.warning(f"{mode.capitalize()} status polling timed out")
+        return False
+
+    def deploy(self, agent_id: str, environment_id: str) -> bool:
+        self._post(f"{self.base_endpoint}/{agent_id}/releases", data={"environment_id": environment_id})
+        return self.poll_release_status(agent_id, environment_id, mode=ReleaseMode.DEPLOY)
+
+    def undeploy(self, agent_id: str, version: str, environment_id: str) -> bool:
+        self._post(f"{self.base_endpoint}/{agent_id}/releases/{version}/undeploy")
+        return self.poll_release_status(agent_id, environment_id, mode=ReleaseMode.UNDEPLOY)
+    
+    def get_environments_for_agent(self, agent_id: str):
+        return self._get(f"{self.base_endpoint}/{agent_id}/environment")
+
     

ibm_watsonx_orchestrate/client/connections/connections_client.py

@@ -19,7 +19,7 @@ class FetchConfigAuthTypes(str, Enum):
     API_KEY_AUTH = ConnectionType.API_KEY_AUTH.value
     OAUTH2_AUTH_CODE = ConnectionType.OAUTH2_AUTH_CODE.value
     OAUTH2_IMPLICIT = 'oauth2_implicit'
-    OAUTH2_PASSWORD = 'oauth2_password'
+    OAUTH2_PASSWORD = ConnectionType.OAUTH2_PASSWORD.value
     OAUTH2_CLIENT_CREDS = ConnectionType.OAUTH2_CLIENT_CREDS.value
     OAUTH_ON_BEHALF_OF_FLOW = ConnectionType.OAUTH_ON_BEHALF_OF_FLOW.value
     KEY_VALUE = ConnectionType.KEY_VALUE.value
@@ -50,7 +50,7 @@ class GetConfigResponse(BaseModel):
 class GetConnectionResponse(BaseModel):
     connection_id: str = None
     app_id: str = None
-    tenant_id: str = None
+    tenant_id: Optional[str] = None
 
 
 
@@ -141,7 +141,8 @@ class ConnectionsClient(BaseAPIClient):
             else:
                 return self._get(f"/connections/applications/runtime_credentials?app_id={app_id}&env={env}")
         except ClientAPIException as e:
-            if e.response.status_code == 404:
+            # Returns 400 when app creds exist but runtime cred don't yet exist
+            if e.response.status_code == 404 or e.response.status_code == 400:
                 return None
             raise e
 

ibm_watsonx_orchestrate/client/knowledge_bases/knowledge_base_client.py

@@ -15,10 +15,10 @@ class KnowledgeBaseClient(BaseAPIClient):
         self.base_endpoint = "/orchestrate/knowledge-bases" if is_local_dev(self.base_url) else "/knowledge-bases"
 
     def create(self, payload: dict) -> dict:
-        return self._post_form_data(f"{self.base_endpoint}/documents", data={ "knowledge_base" : json.dumps(payload) })
+        return self._post_form_data(f"{self.base_endpoint}/documents", data=payload)
     
     def create_built_in(self, payload: dict, files: list) -> dict:
-        return self._post_form_data(f"{self.base_endpoint}/documents", data={ "knowledge_base" : json.dumps(payload) }, files=files)
+        return self._post_form_data(f"{self.base_endpoint}/documents", data=payload, files=files)
 
     def get(self) -> dict:
         return self._get(self.base_endpoint)
@@ -38,10 +38,10 @@ class KnowledgeBaseClient(BaseAPIClient):
         return self._get(f"{self.base_endpoint}/{knowledge_base_id}/status")
 
     def update(self, knowledge_base_id: str, payload: dict) -> dict:
-        return self._patch_form_data(f"{self.base_endpoint}/{knowledge_base_id}/documents", data={ "knowledge_base" : json.dumps(payload) })
+        return self._patch_form_data(f"{self.base_endpoint}/{knowledge_base_id}/documents", data=payload)
     
     def update_with_documents(self, knowledge_base_id: str, payload: dict, files: list) -> dict:
-        return self._patch_form_data(f"{self.base_endpoint}/{knowledge_base_id}/documents", data={ "knowledge_base" : json.dumps(payload) }, files=files)
+        return self._patch_form_data(f"{self.base_endpoint}/{knowledge_base_id}/documents", data=payload, files=files)
 
     def delete(self, knowledge_base_id: str,) -> dict:
         return self._delete(f"{self.base_endpoint}/{knowledge_base_id}")

ibm_watsonx_orchestrate/client/voice_configurations/voice_configurations_client.py

@@ -0,0 +1,75 @@
+from pydantic import ValidationError
+from ibm_watsonx_orchestrate.agent_builder.voice_configurations import VoiceConfiguration
+from ibm_watsonx_orchestrate.client.base_api_client import BaseAPIClient, ClientAPIException
+from ibm_watsonx_orchestrate.client.client_errors import MissingArgument
+
+import logging
+logger = logging.getLogger(__name__)
+
+class VoiceConfigurationsClient(BaseAPIClient):
+
+  def create(self, voice_config: VoiceConfiguration) -> dict:
+    return self._post("/voice_configurations", data=voice_config.model_dump(exclude_none=True))
+
+
+  def update(self, voice_config_id: str, voice_config: VoiceConfiguration) -> dict:    
+    if voice_config_id in [None,""]:
+      raise MissingArgument("voice_config_id")
+    return self._patch(f"/voice_configurations/{voice_config_id}", data=voice_config.model_dump(exclude_none=True))
+
+
+  def get_by_id(self, voice_config_id: str) -> dict | None:
+    if voice_config_id in [None,""]:
+      raise MissingArgument("voice_config_id")
+    
+    try:
+      response = self._get(f"/voice_configurations/{voice_config_id}")
+      return VoiceConfiguration.model_validate(response)
+    
+    except ClientAPIException as e:
+      if e.response.status_code == 404:
+        return None
+      raise e
+    
+    except ValidationError as e:
+      logger.error("Recieved unexpected response from server")
+      raise e
+    
+  def get_by_name(self, name: str) -> list[VoiceConfiguration]:
+    return self.get_by_names([name])
+
+  def get_by_names(self, names: list[str]) -> list[VoiceConfiguration]:
+    # server will implement query by name, below can be uncommented then
+    # formatted_config_names = [f"names={n}" for n in names]
+    # return self._get(f"/voice_configurations?{"&".join(formatted_config_names)}")
+    config_list = self.list()
+    filtered_list = [cfg for cfg in config_list if cfg.name in names]
+
+    try:
+      return [ VoiceConfiguration.model_validate(cfg) for cfg in filtered_list ]
+    except ValidationError as e:
+      logger.error("Recieved unexpected response from server")
+      raise e
+
+
+
+
+  def delete(self, voice_config_id: str) -> None:
+    if voice_config_id in [None,""]:
+      raise MissingArgument("voice_config_id")
+    self._delete(f"/voice_configurations/{voice_config_id}")
+
+
+  def list(self) -> list[dict]:
+    try:
+      response = self._get("/voice_configurations")
+      return [VoiceConfiguration.model_validate(x) for x in response.get('voice_configurations',[])]
+    
+    except ClientAPIException as e:
+      if e.response.status_code == 404:
+        return []
+      raise e
+    
+    except ValidationError as e:
+      logger.error("Recieved unexpected response from server")
+      raise e

ibm_watsonx_orchestrate/docker/compose-lite.yml

@@ -32,6 +32,7 @@ services:
       - wxo-applied-migrations:/var/lib/postgresql/applied_migrations
     command: -c shared_preload_libraries=pgsodium
 
+
   wxo-server-connection-manager:
     image: ${CM_REGISTRY:-us.icr.io/watson-orchestrate-private}/wxo-connections:${CM_TAG:-latest}
     platform: linux/amd64
@@ -102,6 +103,7 @@ services:
       DISMISS_NOTIFICATION_TIMEOUT: 10000
       STANDALONE: "true"
       STREAM_TIMEOUT: ${STREAM_TIMEOUT:-120000}
+      WXO_VOICE_URL: http://localhost:9876/v1
     command: "./StartServer.sh"
     ports:
       - "3000:4002"
@@ -127,8 +129,11 @@ services:
       IBM_DPS_CACHE_SERVICE: https://wxo-doc-processing-cache:8080
       DOCPROC_BASE_URL: http://wxo-doc-processing-infra-standalone:9080
       BUILDER_ASYNC_CALLBACK_ENDPOINT: http://wxo-builder:4025
+      DOCPROC_ENABLED: ${DOCPROC_ENABLED:-false}
       IS_OBSERVABILITY_FEATURE_ENABLED: "true"
       ALLOW_INSECURE_TLS: "true"
+      ENABLE_EDIT_PROMPTS: "true"
+      LANGFLOW_ENABLED: ${LANGFLOW_ENABLED:-false}
     command: 'npm start'
     ports:
       - "4025:4025"
@@ -260,6 +265,7 @@ services:
       - default
       - jaeger-network
     environment:
+      DEFAULT_AGENT_ENABLED: "true"
       RUNTIME_MANAGER_API_KEY: ${RUNTIME_MANAGER_API_KEY:-testapikey} 
       PIP_NO_CACHE_DIR:
       JWT_SECRET: ${JWT_SECRET}
@@ -342,6 +348,8 @@ services:
       AGENTOPS_API_KEY_AUTH_ENABLED: ${AGENTOPS_API_KEY_AUTH_ENABLED:-false}
       AGENTOPS_API_KEY: ${AGENTOPS_API_KEY}
       ES_HOST: http://elasticsearch:9200
+      ORIGIN_HEADER: "internal"
+      LANGFLOW_ENABLED: ${LANGFLOW_ENABLED:-false}
 
   wxo-server-worker:
     image: ${WORKER_REGISTRY:-us.icr.io/watson-orchestrate-private}/wxo-server-conversation_controller:${WORKER_TAG:-latest}
@@ -439,6 +447,7 @@ services:
       USE_IBM_TELEMETRY: ${USE_IBM_TELEMETRY:-false}
       WXO_DEPLOYMENT_PLATFORM: saas
       CALLBACK_HOST_URL: ${CALLBACK_HOST_URL:-http://wxo-server:4321}
+      LANGFLOW_ENABLED: ${LANGFLOW_ENABLED:-false}
   
   tools-runtime-manager:
     image: ${TRM_REGISTRY:-us.icr.io/watson-orchestrate-private}/tools-runtime-manager:${TRM_TAG:-latest}
@@ -480,7 +489,8 @@ services:
       STORAGE_S3_REGION: us-east-1
       AWS_ACCESS_KEY_ID: ${MINIO_ROOT_USER:-minioadmin}
       AWS_SECRET_ACCESS_KEY: ${MINIO_ROOT_PASSWORD:-watsonxorchestrate}
-      TEMPUS_HOST_NAME: http://wxo-tempus-runtime:9044
+      TEMPUS_HOST_NAME: http://wxo-tempus-runtime:9044      
+      LANGFLOW_ENABLED: ${LANGFLOW_ENABLED:-false}
     extra_hosts:
       - "host.docker.internal:host-gateway"
 
@@ -512,12 +522,13 @@ services:
       STORAGE_S3_REGION: us-east-1
       AWS_ACCESS_KEY_ID: ${MINIO_ROOT_USER:-minioadmin}
       AWS_SECRET_ACCESS_KEY: ${MINIO_ROOT_PASSWORD:-watsonxorchestrate}
+      LANGFLOW_ENABLED: ${LANGFLOW_ENABLED:-false}
 
   ########################
   # IBM AGENT-OPS
   ########################
   elasticsearch:
-    image: docker.elastic.co/elasticsearch/elasticsearch:8.17.5
+    image: docker.elastic.co/elasticsearch/elasticsearch:8.19.0
     profiles: [ibm-telemetry]
     environment:
       - discovery.type=single-node
@@ -790,10 +801,12 @@ services:
       LOG_LEVEL: info
       DOCPROC_ENABLED: ${DOCPROC_ENABLED:-false}
       DOCPROC_BASE_URL: http://wxo-doc-processing-infra-standalone:9080
+      DOCPROC_OUTPUT_FILE_SERVER_URL: ${DOCPROC_OUTPUT_FILE_SERVER_URL:-http://wxo-server:4321}
       WO_API_KEY: ${WO_API_KEY}
       WO_INSTANCE: ${WO_INSTANCE}
       AUTHORIZATION_URL: ${AUTHORIZATION_URL}
       WO_AUTH_TYPE: ${WO_AUTH_TYPE}
+      FLOW_CALLBACK_REQUEST_RETRIES: "1"
     healthcheck:
       test: curl -k http://localhost:9044/readiness --fail
       interval: 30s
@@ -869,6 +882,19 @@ services:
       PREFLIGHT_MAX_SIZE: 10Mb
       PREFLIGHT_MAX_PAGES: 600
       RUNTIME_LIBRARY: "watson_doc_understanding"
+      WXAI_API_KEY: ${WXAI_API_KEY:-}
+      WXAI_IMAGE_DESCRIPTION_MODEL_ID: ${WXAI_IMAGE_DESCRIPTION_MODEL_ID:-mistralai/pixtral-12b}
+      WXAI_IMAGE_DESCRIPTION_PROJECT_ID: ${WXAI_IMAGE_DESCRIPTION_PROJECT_ID:-}
+      WXAI_KVP_MODEL_ID: ${WXAI_KVP_MODEL_ID:-}
+      WXAI_KVP_PROJECT_ID: ${WXAI_KVP_PROJECT_ID:-}
+      WXAI_SEMANTIC_KVP_PROJECT_ID: ${WXAI_SEMANTIC_KVP_PROJECT_ID:-}
+      WXAI_SEMANTIC_KVP_MODEL_ID: ${WXAI_SEMANTIC_KVP_MODEL_ID:-mistralai/pixtral-12b}
+      WXAI_SEMANTIC_KVP_SPACE_ID: ${WXAI_SEMANTIC_KVP_SPACE_ID:-}
+      WXAI_URL: ${WXAI_URL:-}
+      WXAI_USERNAME: ${WXAI_USERNAME:-}
+      WXAI_INSTANCE_ID: ${WXAI_INSTANCE_ID:-}
+      WXAI_VERSION: ${WXAI_VERSION:-5.1}
+      
     profiles:
       - docproc
     ports:
@@ -997,9 +1023,7 @@ services:
       SERVICE_NAME: wxo-doc-processing-llm-service
       LOG_LEVEL: info
       WATSONX_API_ENDPOINT: ${WATSONX_URL:-https://us-south.ml.cloud.ibm.com}
-      WATSONX_API_KEY: ${WATSONX_APIKEY}
-      WATSONX_PROJECT_ID: ${WATSONX_PROJECT_ID}
-      WATSONX_SPACE_ID: ${WATSONX_SPACE_ID}
+      WXO_SERVER_BASE_URL : http://wxo-server:4321
       IBM_DPS_CACHE_SERVICE: 'wxo-doc-processing-cache:8080'
     profiles:
       - docproc
@@ -1092,6 +1116,30 @@ services:
       exit 0;
       "
 
+  wxo-server-voice:
+    image: ${VOICE_CONTROLLER_REGISTRY:-us.icr.io/watson-orchestrate-private}/wxo-server-voice:${VOICE_CONTROLLER_TAG:-latest}
+    profiles: [voice]
+    restart: unless-stopped
+    ports: 
+      - 9876:8765
+    environment:
+      WXO_BASE_DOMAIN_URL: "http://wxo-server:4321"
+      AI_GATEWAY_URL: "http://ai-gateway:8787"
+      WXO_ORIGIN_HEADER: "internal"
+
+  langflow:
+    image: ${OPENSOURCE_REGISTRY_PROXY:-docker.io}/langflowai/${LANGFLOW_IMAGE:-langflow}:${LANGFLOW_TAG:-latest}
+    profiles: [langflow]
+    ports:
+      - 7861:7861
+    environment:
+      LANGFLOW_PORT: 7861
+      LANGFLOW_DATABASE_URL: postgresql://${POSTGRES_USER:-postgres}:${POSTGRES_PASSWORD:-postgres}@wxo-server-db:5432/langflow
+      LANGFLOW_CONFIG_DIR: app/langflow
+    volumes:
+      - langflow-data:/app/langflow
+
+
 volumes:
   tools:
     driver: local
@@ -1124,6 +1172,7 @@ volumes:
   wdu-models:
     driver: local
   es_data:
+  langflow-data:
 
 networks:
   default:

ibm_watsonx_orchestrate/docker/default.env

@@ -57,13 +57,14 @@ EVENT_BROKER_TTL="3600"
 REGISTRY_URL=
 
 
-SERVER_TAG=01-08-2025
+
+SERVER_TAG=20-08-2025-cf6e342
 SERVER_REGISTRY=
 
-WORKER_TAG=01-08-2025-v2
+WORKER_TAG=20-08-2025-cf6e342
 WORKER_REGISTRY=
 
-AI_GATEWAY_TAG=01-08-2025-v1
+AI_GATEWAY_TAG=20-08-2025-9ed6d40
 AI_GATEWAY_REGISTRY=
 
 AGENT_GATEWAY_TAG=29-07-2025
@@ -82,16 +83,16 @@ UITAG=31-07-2025
 CM_REGISTRY=
 CM_TAG=24-07-2025
 
-TRM_TAG=23-07-2025-3c60549f0bac275de3e5736265a3fd49cdd3a203
+TRM_TAG=19-08-2025-fe105eb0b950ff304f712a1a5b9fa3cba92d09da
 TRM_REGISTRY=
 
-TR_TAG=23-07-2025-3c60549f0bac275de3e5736265a3fd49cdd3a203
+TR_TAG=19-08-2025-fe105eb0b950ff304f712a1a5b9fa3cba92d09da
 TR_REGISTRY=
 
-BUILDER_TAG=31-07-2025-d7145cb
+BUILDER_TAG=19-08-2025-1a79d34
 BUILDER_REGISTRY=
 
-FLOW_RUNTIME_TAG=01-08-2025
+FLOW_RUNTIME_TAG=18-08-2025-v2
 FLOW_RUMTIME_REGISTRY=
 
 
@@ -104,17 +105,23 @@ JAEGER_PROXY_REGISTRY=
 SOCKET_HANDLER_TAG=29-05-2025
 SOCKET_HANDLER_REGISTRY=
 
-CPE_TAG=06-08-2025-b0a20ad
+CPE_TAG=18-08-2025-ae1308e
 CPE_REGISTRY=
 
+VOICE_CONTROLLER_TAG=12-08-2025
+VOICE_CONTROLLER_REGISTRY=
+
+LANGFLOW_TAG=
+LANGFLOW_IMAGE=
+
 # IBM Document Processing
-WDU_TAG=2.5.0
+WDU_TAG=2.6.1
 WDU_REGISTRY=
 
-DOCPROC_DPS_TAG=20250721-164412-250-503756a
-DOCPROC_LLMSERVICE_TAG=20250725-100249-111-51d3e51
-DOCPROC_CACHE_TAG=20250723-100852-70-9edc1ab
-DOCPROC_DPI_TAG=20250731-155328-257-06879e86
+DOCPROC_DPS_TAG=20250815-010747-277-173db2a
+DOCPROC_LLMSERVICE_TAG=20250820-153924-128-55cf4d5
+DOCPROC_CACHE_TAG=20250814-master-82-cf33f87
+DOCPROC_DPI_TAG=20250815-004755-273-e65f26b4
 DOCPROC_REGISTRY=
 
 # END -- IMAGE REGISTRIES AND TAGS
@@ -149,6 +156,7 @@ TOOLS_RUNTIME_MANAGER_BASE_URL="http://tools-runtime-manager:8080"
 CONNECTION_SERVICE_BASE_URL="http://wxo-server-connection-manager:3001"
 AI_GATEWAY_BASE_URL="http://ai-gateway:8787/v1"
 AI_GATEWAY_ENABLED=True
+DEFAULT_AGENT_ENABLED=True
 AGENT_GATEWAY_URI="http://wxo-agent-gateway:8989"
 DEFAULT_TENANT_ID=10000000-0000-0000-0000-000000000000
 ES_USERNAME=elastic

ibm_watsonx_orchestrate/flow_builder/flows/__init__.py

@@ -19,6 +19,8 @@ __all__ = [
     "DocProcInput",
 
     "DocProcNode",
+    "DocExtNode",
+    "DocClassifierNode"
     "UserNode",
     "AgentNode",
     "StartNode",