agentex-sdk 0.6.1__tar.gz → 0.6.3__tar.gz

agentex_sdk-0.6.3/.release-please-manifest.json

@@ -0,0 +1,3 @@
+{
+  ".": "0.6.3"
+}

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/CHANGELOG.md

@@ -1,5 +1,17 @@
 # Changelog
 
+## 0.6.3 (2025-11-06)
+
+Full Changelog: [v0.6.2...v0.6.3](https://github.com/scaleapi/scale-agentex-python/compare/v0.6.2...v0.6.3)
+
+## 0.6.2 (2025-11-05)
+
+Full Changelog: [v0.6.1...v0.6.2](https://github.com/scaleapi/scale-agentex-python/compare/v0.6.1...v0.6.2)
+
+### Features
+
+* **api:** update via SDK Studio ([b732dfa](https://github.com/scaleapi/scale-agentex-python/commit/b732dfac50cacc90c84a751fd6c75d18fa5b43ed))
+
 ## 0.6.1 (2025-11-05)
 
 Full Changelog: [v0.6.0...v0.6.1](https://github.com/scaleapi/scale-agentex-python/compare/v0.6.0...v0.6.1)

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: agentex-sdk
-Version: 0.6.1
+Version: 0.6.3
 Summary: The official Python library for the agentex API
 Project-URL: Homepage, https://github.com/scaleapi/scale-agentex-python
 Project-URL: Repository, https://github.com/scaleapi/scale-agentex-python

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/api.md

@@ -152,3 +152,16 @@ Methods:
 - <code title="get /tracker/{tracker_id}">client.tracker.<a href="./src/agentex/resources/tracker.py">retrieve</a>(tracker_id) -> <a href="./src/agentex/types/agent_task_tracker.py">AgentTaskTracker</a></code>
 - <code title="put /tracker/{tracker_id}">client.tracker.<a href="./src/agentex/resources/tracker.py">update</a>(tracker_id, \*\*<a href="src/agentex/types/tracker_update_params.py">params</a>) -> <a href="./src/agentex/types/agent_task_tracker.py">AgentTaskTracker</a></code>
 - <code title="get /tracker">client.tracker.<a href="./src/agentex/resources/tracker.py">list</a>(\*\*<a href="src/agentex/types/tracker_list_params.py">params</a>) -> <a href="./src/agentex/types/tracker_list_response.py">TrackerListResponse</a></code>
+
+# DeploymentHistory
+
+Types:
+
+```python
+from agentex.types import DeploymentHistory, DeploymentHistoryListResponse
+```
+
+Methods:
+
+- <code title="get /deployment-history/{deployment_id}">client.deployment_history.<a href="./src/agentex/resources/deployment_history.py">retrieve</a>(deployment_id) -> <a href="./src/agentex/types/deployment_history.py">DeploymentHistory</a></code>
+- <code title="get /deployment-history">client.deployment_history.<a href="./src/agentex/resources/deployment_history.py">list</a>(\*\*<a href="src/agentex/types/deployment_history_list_params.py">params</a>) -> <a href="./src/agentex/types/deployment_history_list_response.py">DeploymentHistoryListResponse</a></code>

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/00_sync/000_hello_acp/manifest.yaml

@@ -61,7 +61,7 @@ agent:
 
   # Type of ACP to use
   # sync: Simple synchronous ACP implementation
-  # agentic: Advanced ACP with sub-types "base" or "temporal" (requires config)
+  # async: Asynchronous, non-blocking ACP implementation
   acp_type: sync
 
   # Description of what your agent does

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/000_hello_acp/README.md

@@ -44,6 +44,6 @@ Three handlers instead of one, giving you full control over task lifecycle. Task
 - Building towards production systems
 
 ## Why This Matters
-The task-based model is the foundation of production agents. Unlike sync agents where each message is independent, agentic agents maintain persistent tasks that can receive multiple events, store state, and have full lifecycle management. This is the stepping stone to Temporal-based agents.
+The task-based model is the foundation of production agents. Unlike sync agents where each message is independent, async agents maintain persistent tasks that can receive multiple events, store state, and have full lifecycle management. This is the stepping stone to Temporal-based agents.
 
 **Next:** [010_multiturn](../010_multiturn/) - Add conversation memory

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/000_hello_acp/manifest.yaml

@@ -106,7 +106,7 @@ deployment:
   global:
     agent:
       name: "ab000-hello-acp"
-      description: "An AgentEx agent that is not intelligent. It just shows how to implement the base agentic ACP type."
+      description: "An AgentEx agent that is not intelligent. It just shows how to implement the base async ACP type."
     
     # Default replica count
     replicaCount: 1

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/010_multiturn/README.md

@@ -1,27 +1,27 @@
-# [Agentic] Multiturn
+# [Async] Multiturn
 
-Handle multi-turn conversations in agentic agents with task-based state management. Each task maintains its own conversation history automatically.
+Handle multi-turn conversations in async agents with task-based state management. Each task maintains its own conversation history automatically.
 
 ## What You'll Learn
 - How tasks maintain conversation state across multiple exchanges
-- Difference between sync and agentic multiturn patterns
+- Difference between sync and async multiturn patterns
 - Building stateful conversational agents with minimal code
 
 ## Prerequisites
 - Development environment set up (see [main repo README](https://github.com/scaleapi/scale-agentex))
 - Backend services running: `make dev` from repository root
-- Understanding of basic agentic agents (see [000_hello_acp](../000_hello_acp/))
+- Understanding of basic async agents (see [000_hello_acp](../000_hello_acp/))
 
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/00_base/010_multiturn
+cd examples/tutorials/10_async/00_base/010_multiturn
 uv run agentex agents run --manifest manifest.yaml
 ```
 
 ## Key Pattern
 
-Unlike sync agents where you manually track conversation history, agentic agents automatically maintain state within each task:
+Unlike sync agents where you manually track conversation history, async agents automatically maintain state within each task:
 
 ```python
 @app.on_task_event_send()

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/010_multiturn/dev.ipynb

@@ -29,7 +29,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# (REQUIRED) Create a new task. For Agentic agents, you must create a task for messages to be associated with.\n",
+    "# (REQUIRED) Create a new task. For Async agents, you must create a task for messages to be associated with.\n",
     "import uuid\n",
     "\n",
     "rpc_response = client.agents.create_task(\n",

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/020_streaming/README.md

@@ -1,21 +1,21 @@
 # [Agentic] Streaming
 
-Stream responses in agentic agents using `adk.messages.create()` to send progressive updates. More flexible than sync streaming since you can send multiple messages at any time.
+Stream responses in async agents using `adk.messages.create()` to send progressive updates. More flexible than sync streaming since you can send multiple messages at any time.
 
 ## What You'll Learn
 - How to stream with explicit message creation
-- Difference between sync and agentic streaming patterns
+- Difference between sync and async streaming patterns
 - When to send multiple messages vs single streamed response
 
 ## Prerequisites
 - Development environment set up (see [main repo README](https://github.com/scaleapi/scale-agentex))
 - Backend services running: `make dev` from repository root
-- Understanding of agentic basics (see [000_hello_acp](../000_hello_acp/))
+- Understanding of async basics (see [000_hello_acp](../000_hello_acp/))
 
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/00_base/020_streaming
+cd examples/tutorials/10_async/00_base/020_streaming
 uv run agentex agents run --manifest manifest.yaml
 ```
 
@@ -33,7 +33,7 @@ async def handle_event_send(params: SendEventParams):
     await adk.messages.create(task_id=task_id, content=...)
 ```
 
-Unlike sync streaming (which uses async generators), agentic streaming uses explicit message creation calls, giving you more control over when and what to send.
+Unlike sync streaming (which uses async generators), async streaming uses explicit message creation calls, giving you more control over when and what to send.
 
 ## When to Use
 - Multi-step processes with intermediate results

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/030_tracing/README.md

@@ -11,12 +11,12 @@ Add observability to your agents with spans and traces using `adk.tracing.start_
 ## Prerequisites
 - Development environment set up (see [main repo README](https://github.com/scaleapi/scale-agentex))
 - Backend services running: `make dev` from repository root
-- Understanding of agentic agents (see [000_hello_acp](../000_hello_acp/))
+- Understanding of async agents (see [000_hello_acp](../000_hello_acp/))
 
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/00_base/030_tracing
+cd examples/tutorials/10_async/00_base/030_tracing
 uv run agentex agents run --manifest manifest.yaml
 ```
 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/040_other_sdks/README.md

@@ -11,12 +11,12 @@ Agents are just Python code - integrate any SDK you want (OpenAI, Anthropic, Lan
 ## Prerequisites
 - Development environment set up (see [main repo README](https://github.com/scaleapi/scale-agentex))
 - Backend services running: `make dev` from repository root
-- Understanding of agentic agents (see [000_hello_acp](../000_hello_acp/))
+- Understanding of async agents (see [000_hello_acp](../000_hello_acp/))
 
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/00_base/040_other_sdks
+cd examples/tutorials/10_async/00_base/040_other_sdks
 uv run agentex agents run --manifest manifest.yaml
 ```
 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/080_batch_events/README.md

@@ -1,9 +1,9 @@
 # [Agentic] Batch Events
 
-Demonstrates limitations of the base agentic protocol with concurrent event processing. When multiple events arrive rapidly, base agentic agents handle them sequentially, which can cause issues.
+Demonstrates limitations of the base async protocol with concurrent event processing. When multiple events arrive rapidly, base async agents handle them sequentially, which can cause issues.
 
 ## What You'll Learn
-- Limitations of non-Temporal agentic agents
+- Limitations of non-Temporal async agents
 - Race conditions and ordering issues in concurrent scenarios
 - When you need workflow orchestration
 - Why this motivates Temporal adoption
@@ -11,12 +11,12 @@ Demonstrates limitations of the base agentic protocol with concurrent event proc
 ## Prerequisites
 - Development environment set up (see [main repo README](https://github.com/scaleapi/scale-agentex))
 - Backend services running: `make dev` from repository root
-- Understanding of agentic patterns (see previous tutorials)
+- Understanding of async patterns (see previous tutorials)
 
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/00_base/080_batch_events
+cd examples/tutorials/10_async/00_base/080_batch_events
 uv run agentex agents run --manifest manifest.yaml
 ```
 
@@ -27,7 +27,7 @@ This tutorial shows **when you need Temporal**. If your agent needs to:
 - Process multiple events in parallel safely
 - Maintain consistent state under concurrent load
 
-Then you should use Temporal workflows (see tutorials 10_agentic/10_temporal/) which provide:
+Then you should use Temporal workflows (see tutorials 10_async/10_temporal/) which provide:
 - Deterministic event ordering
 - Safe concurrent processing
 - Guaranteed state consistency
@@ -35,7 +35,7 @@ Then you should use Temporal workflows (see tutorials 10_agentic/10_temporal/) w
 This is the "breaking point" tutorial that motivates moving to Temporal for production agents.
 
 ## When to Use (This Pattern)
-This tutorial shows what NOT to use for production. Use base agentic agents only when:
+This tutorial shows what NOT to use for production. Use base async agents only when:
 - Events are infrequent (< 1 per second)
 - Order doesn't matter
 - State consistency isn't critical

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/090_multi_agent_non_temporal/README.md

@@ -49,13 +49,13 @@ The system uses a shared build configuration with type-safe interfaces:
 - Backend services running: `make dev` from repository root
 - Python 3.12+ and uv package manager
 - OpenAI API key (set `OPENAI_API_KEY` or create `.env` file)
-- Understanding of agentic patterns (see previous tutorials)
+- Understanding of async patterns (see previous tutorials)
 
 ### Running the System
 
 1. **Start all agents**:
    ```bash
-   cd examples/tutorials/10_agentic/00_base/090_multi_agent_non_temporal
+   cd examples/tutorials/10_async/00_base/090_multi_agent_non_temporal
    ./start-agents.sh start
    ```
 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/00_base/090_multi_agent_non_temporal/start-agents.sh

@@ -18,7 +18,7 @@ CRITIC_PORT=8002
 FORMATTER_PORT=8003
 
 # Base directory
-BASE_DIR="examples/tutorials/10_agentic/00_base/090_multi_agent_non_temporal"
+BASE_DIR="examples/tutorials/10_async/00_base/090_multi_agent_non_temporal"
 
 echo -e "${BLUE}🎭 Multi-Agent Content Assembly Line (Flattened)${NC}"
 echo -e "${BLUE}===============================================${NC}"

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/000_hello_acp/README.md

@@ -6,18 +6,18 @@ Temporal workflows make agents durable - they survive restarts and can run indef
 - Building durable agents with Temporal workflows
 - The workflow and signal pattern
 - How workflows survive failures and resume automatically
-- When to use Temporal vs base agentic agents
+- When to use Temporal vs base async agents
 
 ## Prerequisites
 - Development environment set up (see [main repo README](https://github.com/scaleapi/scale-agentex))
 - Backend services running: `make dev` from repository root (includes Temporal)
 - Temporal UI available at http://localhost:8233
-- Understanding of base agentic agents (see [../../00_base/080_batch_events](../../00_base/080_batch_events/) to understand why Temporal)
+- Understanding of base async agents (see [../../00_base/080_batch_events](../../00_base/080_batch_events/) to understand why Temporal)
 
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/10_temporal/000_hello_acp
+cd examples/tutorials/10_async/10_temporal/000_hello_acp
 uv run agentex agents run --manifest manifest.yaml
 ```
 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/010_agent_chat/README.md

@@ -17,7 +17,7 @@ Combine streaming responses, multi-turn chat, tool calling, and tracing - all wi
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/10_temporal/010_agent_chat
+cd examples/tutorials/10_async/10_temporal/010_agent_chat
 uv run agentex agents run --manifest manifest.yaml
 ```
 
@@ -33,7 +33,7 @@ uv run agentex agents run --manifest manifest.yaml
 
 ## Key Insight
 
-In base agentic agents, all this state lives in memory and is lost on crash. With Temporal, the entire conversation - history, tool calls, intermediate state - is durably persisted. The agent can pick up a conversation that paused days ago as if no time passed.
+In base async agents, all this state lives in memory and is lost on crash. With Temporal, the entire conversation - history, tool calls, intermediate state - is durably persisted. The agent can pick up a conversation that paused days ago as if no time passed.
 
 ## When to Use
 - Production chatbots with tool capabilities

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/010_agent_chat/project/workflow.py

@@ -234,7 +234,7 @@ class At010AgentChatWorkflow(BaseWorkflow):
                     "to provide accurate and well-reasoned responses."
                 ),
                 parent_span_id=span.id if span else None,
-                model="gpt-5-mini",
+                model="gpt-4o-mini",
                 model_settings=ModelSettings(
                     # Include reasoning items in the response (IDs, summaries)
                     # response_include=["reasoning.encrypted_content"],

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/010_agent_chat/tests/test_agent.py

@@ -7,7 +7,7 @@ This test suite demonstrates how to test agents that integrate:
 - Multi-turn conversations with state management
 - Tool usage (calculator and web search via MCP)
 
-Key differences from base agentic (040_other_sdks):
+Key differences from base async (040_other_sdks):
 1. Temporal Integration: Uses Temporal workflows for durable execution
 2. State Management: State is managed within the workflow instance
 3. No Race Conditions: Temporal ensures sequential event processing

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/020_state_machine/README.md

@@ -17,7 +17,7 @@ Build complex multi-state workflows using state machines with Temporal. This tut
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/10_temporal/020_state_machine
+cd examples/tutorials/10_async/10_temporal/020_state_machine
 uv run agentex agents run --manifest manifest.yaml
 ```
 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/020_state_machine/manifest.yaml

@@ -73,7 +73,7 @@ agent:
 
   # Description of what your agent does
   # Helps with documentation and discovery
-  description: An AgentEx agentthat demonstrates how to uose state machines to manage complex agentic workflows
+  description: An AgentEx agentthat demonstrates how to uose state machines to manage complex async workflows
 
   # Temporal workflow configuration
   # This enables your agent to run as a Temporal workflow for long-running tasks
@@ -122,7 +122,7 @@ deployment:
   global:
     agent:
       name: "at020-state-machine"
-      description: "An AgentEx agentthat demonstrates how to uose state machines to manage complex agentic workflows"
+      description: "An AgentEx agentthat demonstrates how to uose state machines to manage complex async workflows"
     
     # Default replica count
     replicaCount: 1

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/020_state_machine/pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "hatchling.build"
 [project]
 name = "at020-state-machine"
 version = "0.1.0"
-description = "An AgentEx agentthat demonstrates how to uose state machines to manage complex agentic workflows"
+description = "An AgentEx agentthat demonstrates how to uose state machines to manage complex async workflows"
 readme = "README.md"
 requires-python = ">=3.12"
 dependencies = [

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/030_custom_activities/README.md

@@ -18,7 +18,7 @@ Learn how to extend Temporal workflows with custom activities for external opera
 
 **Terminal 1 - Start Worker:**
 ```bash
-cd examples/tutorials/10_agentic/10_temporal/030_custom_activities
+cd examples/tutorials/10_async/10_temporal/030_custom_activities
 uv run python project/run_worker.py
 ```
 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/050_agent_chat_guardrails/README.md

@@ -17,7 +17,7 @@ This tutorial demonstrates how to implement streaming multiturn tool-enabled cha
 ## Quick Start
 
 ```bash
-cd examples/tutorials/10_agentic/10_temporal/050_agent_chat_guardrails
+cd examples/tutorials/10_async/10_temporal/050_agent_chat_guardrails
 uv run agentex agents run --manifest manifest.yaml
 ```
 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/050_agent_chat_guardrails/project/acp.py

@@ -5,7 +5,7 @@ from agentex.lib.sdk.fastacp.fastacp import FastACP
 
 # Create the ACP server
 acp = FastACP.create(
-    acp_type="agentic",
+    acp_type="async",
     config=TemporalACPConfig(
         # When deployed to the cluster, the Temporal address will automatically be set to the cluster address
         # For local development, we set the address manually to talk to the local Temporal service set up via docker compose

{agentex_sdk-0.6.1/examples/tutorials/10_async/10_temporal/070_open_ai_agents_sdk_tools → agentex_sdk-0.6.3/examples/tutorials/10_async/10_temporal/060_open_ai_agents_sdk_hello_world}/dev.ipynb

@@ -18,9 +18,7 @@
    "id": "d1c309d6",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "AGENT_NAME = \"example-tutorial\""
-   ]
+   "source": "AGENT_NAME = \"at060-open-ai-agents-sdk-hello-world\""
   },
   {
    "cell_type": "code",

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/060_open_ai_agents_sdk_hello_world/manifest.yaml

@@ -69,7 +69,7 @@ agent:
 
   # Unique name for your agent
   # Used for task routing and monitoring
-  name: example-tutorial
+  name: at060-open-ai-agents-sdk-hello-world
 
   # Description of what your agent does
   # Helps with documentation and discovery
@@ -82,12 +82,12 @@ agent:
     workflows:
       # Name of the workflow class
       # Must match the @workflow.defn name in your workflow.py
-      - name: example-tutorial
+      - name: at060-open-ai-agents-sdk-hello-world
 
         # Queue name for task distribution
         # Used by Temporal to route tasks to your agent
         # Convention: <agent_name>_task_queue
-        queue_name: example_tutorial_queue
+        queue_name: at060_open_ai_agents_sdk_hello_world_queue
 
   # Optional: Credentials mapping
   # Maps Kubernetes secrets to environment variables
@@ -96,9 +96,9 @@ agent:
     - env_var_name: REDIS_URL
       secret_name: redis-url-secret
       secret_key: url
-  #   - env_var_name: OPENAI_API_KEY
-  #     secret_name: openai-api-key
-  #     secret_key: api-key
+    # - env_var_name: OPENAI_API_KEY
+    #   secret_name: openai-api-key
+    #   secret_key: api-key
   
   # Optional: Set Environment variables for running your agent locally as well 
   # as for deployment later on
@@ -124,7 +124,7 @@ deployment:
   # These can be overridden using --override-file with custom configuration files
   global:
     agent:
-      name: "example-tutorial"
+      name: "at060-open-ai-agents-sdk-hello-world"
       description: "An AgentEx agent"
     
     # Default replica count
@@ -137,4 +137,4 @@ deployment:
         memory: "1Gi"
       limits:
         cpu: "1000m"
-        memory: "2Gi" 
+        memory: "2Gi" 

{agentex_sdk-0.6.1 → agentex_sdk-0.6.3}/examples/tutorials/10_async/10_temporal/060_open_ai_agents_sdk_hello_world/project/acp.py

@@ -34,6 +34,13 @@ if os.getenv("AGENTEX_DEBUG_ENABLED") == "true":
 
 from agentex.lib.types.fastacp import TemporalACPConfig
 from agentex.lib.sdk.fastacp.fastacp import FastACP
+from agentex.lib.core.temporal.plugins.openai_agents.models.temporal_streaming_model import (
+    TemporalStreamingModelProvider,
+)
+from agentex.lib.core.temporal.plugins.openai_agents.interceptors.context_interceptor import ContextInterceptor
+
+context_interceptor = ContextInterceptor()
+temporal_streaming_model_provider = TemporalStreamingModelProvider()
 
 # Create the ACP server
 acp = FastACP.create(
@@ -44,7 +51,8 @@ acp = FastACP.create(
         # We are also adding the Open AI Agents SDK plugin to the ACP.
         type="temporal",
         temporal_address=os.getenv("TEMPORAL_ADDRESS", "localhost:7233"),
-        plugins=[OpenAIAgentsPlugin()]
+        plugins=[OpenAIAgentsPlugin(model_provider=temporal_streaming_model_provider)],
+        interceptors=[context_interceptor]
     )
 )
 

agentex_sdk-0.6.3/examples/tutorials/10_async/10_temporal/060_open_ai_agents_sdk_hello_world/project/run_worker.py

@@ -0,0 +1,69 @@
+import asyncio
+
+from temporalio.contrib.openai_agents import OpenAIAgentsPlugin
+
+from project.workflow import At060OpenAiAgentsSdkHelloWorldWorkflow
+from agentex.lib.utils.debug import setup_debug_if_enabled
+from agentex.lib.utils.logging import make_logger
+from agentex.lib.environment_variables import EnvironmentVariables
+from agentex.lib.core.temporal.activities import get_all_activities
+from agentex.lib.core.temporal.workers.worker import AgentexWorker
+from agentex.lib.core.temporal.plugins.openai_agents.models.temporal_streaming_model import (
+    TemporalStreamingModelProvider,
+)
+from agentex.lib.core.temporal.plugins.openai_agents.interceptors.context_interceptor import ContextInterceptor
+
+environment_variables = EnvironmentVariables.refresh()
+
+logger = make_logger(__name__)
+
+
+async def main():
+    # Setup debug mode if enabled
+    setup_debug_if_enabled()
+    
+    task_queue_name = environment_variables.WORKFLOW_TASK_QUEUE
+    if task_queue_name is None:
+        raise ValueError("WORKFLOW_TASK_QUEUE is not set")
+    
+    # Add activities to the worker
+    all_activities = get_all_activities() + []  # add your own activities here
+
+    # ============================================================================
+    # STREAMING SETUP: Interceptor + Model Provider
+    # ============================================================================
+    # This is where the streaming magic is configured! Two key components:
+    #
+    # 1. ContextInterceptor
+    #    - Threads task_id through activity headers using Temporal's interceptor pattern
+    #    - Outbound: Reads _task_id from workflow instance, injects into activity headers
+    #    - Inbound: Extracts task_id from headers, sets streaming_task_id ContextVar
+    #    - This enables runtime context without forking the Temporal plugin!
+    #
+    # 2. TemporalStreamingModelProvider
+    #    - Returns TemporalStreamingModel instances that read task_id from ContextVar
+    #    - TemporalStreamingModel.get_response() streams tokens to Redis in real-time
+    #    - Still returns complete response to Temporal for determinism/replay safety
+    #    - Uses AgentEx ADK streaming infrastructure (Redis XADD to stream:{task_id})
+    #
+    # Together, these enable real-time LLM streaming while maintaining Temporal's
+    # durability guarantees. No forked components - uses STANDARD OpenAIAgentsPlugin!
+    context_interceptor = ContextInterceptor()
+    temporal_streaming_model_provider = TemporalStreamingModelProvider()
+
+    # Create a worker with automatic tracing
+    # IMPORTANT: We use the STANDARD temporalio.contrib.openai_agents.OpenAIAgentsPlugin
+    # No forking needed! The interceptor + model provider handle all streaming logic.
+    worker = AgentexWorker(
+        task_queue=task_queue_name,
+        plugins=[OpenAIAgentsPlugin(model_provider=temporal_streaming_model_provider)],
+        interceptors=[context_interceptor]
+    )
+
+    await worker.run(
+        activities=all_activities,
+        workflow=At060OpenAiAgentsSdkHelloWorldWorkflow,
+    )
+
+if __name__ == "__main__":
+    asyncio.run(main())