Temporal Python SDK Samples

This is a collection of samples showing how to use the Python SDK.

Usage

Prerequisites:

• uv
• Temporal CLI with a local dev server running:

temporal server start-dev

The SDK requires Python >= 3.10. You can install Python using uv. For example,

uv python install 3.13

With this repository cloned, run the following at the root of the directory:

uv sync

That loads all required dependencies. Then to run a sample, usually you just run it under uv. For example:

uv run hello/hello_activity.py

Some examples require extra dependencies. See each sample's directory for specific instructions.

Samples

• hello - All of the basic features.

• activity_worker - Use Python activities from a workflow in another language.
• batch_sliding_window - Batch processing with a sliding window of child workflows.
• bedrock - Orchestrate a chatbot with Amazon Bedrock.
• bedrock_agentcore/strands_agent - Run a AWS Strands Agent with Temporal Plugin on AgentCore Worker.
• cloud_export_to_parquet - Set up schedule workflow to process exported files on an hourly basis
• context_propagation - Context propagation through workflows/activities via interceptor.
• custom_converter - Use a custom payload converter to handle custom types.
• custom_decorator - Custom decorator to auto-heartbeat a long-running activity.
• custom_metric - Custom metric to record the workflow type in the activity schedule to start latency.
• deepagents_plugin - Make LangChain Deep Agents durable: each LLM/tool/backend call becomes a Temporal Activity while the agent loop replays in the Workflow.
• dsl - DSL workflow that executes steps defined in a YAML file.
• eager_wf_start - Run a workflow using Eager Workflow Start
• encryption - Apply end-to-end encryption for all input/output.
• env_config - Load client configuration from TOML files with programmatic overrides.
• external_storage - Offload large payloads to S3-compatible object storage, plus a codec server for the Web UI and CLI.
• external_storage_redis - Redis driver for external storage
• gevent_async - Combine gevent and Temporal.
• google_adk_agents - Run Google ADK agents as durable Temporal workflows (model calls, tools, multi-agent, MCP, streaming).
• google_genai - Run the Google Gemini SDK inside durable Temporal workflows.
• hello_nexus - Define a Nexus service, implement operation handlers, and call them from a workflow.
• hello_standalone_nexus - Use Nexus Operations without using a workflow.
• hello_standalone_activity - Use activities without using a workflow.
• lambda_worker - Run a Temporal Worker inside an AWS Lambda function.
• langfuse_tracing - Trace Temporal workflows in Langfuse with the OpenTelemetry plugin and OTLP export.
• langgraph_plugin - Run LangGraph workflows as durable Temporal workflows (Graph API and Functional API).
• langsmith_tracing - Trace Temporal workflows with LangSmith via the LangSmith plugin.
• litellm_activity - Call LLM providers through LiteLLM from a Temporal Activity.
• message_passing/introduction - Introduction to queries, signals, and updates.
• message_passing/safe_message_handlers - Safely handling updates and signals.
• message_passing/update_with_start/lazy_initialization - Use update-with-start to update a Shopping Cart, starting it if it does not exist.
• nexus_cancel - Fan out concurrent Nexus operations, take the first result, and cancel the rest.
• nexus_messaging: Demonstrates how to send signal, update and query messages through Nexus.

• nexus_messaging_temporal_operation: Demonstrates how to send signal, update and query messages through Nexus.

• nexus_multiple_args - Map a Nexus operation to a handler workflow that takes multiple arguments.
• nexus_standalone_activity - Back a Nexus operation with a standalone Activity.
• nexus_standalone_operations - Execute Nexus operations directly from client code,

• open_telemetry - Trace workflows with OpenTelemetry.
• openai_agents - Run OpenAI Agents SDK agents as durable Temporal workflows.
• openrouter - Call OpenRouter from Activities: fan out a prompt batch, and pause instead of failing when the budget or credits run out.
• patching - Alter workflows safely with patch and deprecate_patch.
• polling - Recommended implementation of an activity that needs to periodically poll an external resource waiting its successful completion.
• prometheus - Configure Prometheus metrics on clients/workers.
• pydantic_converter - Data converter for using Pydantic models.
• pydantic_converter_v1 - Data converter for Pydantic v1 models (prefer pydantic_converter for v2).
• replay - Verify that workflow code changes are compatible with existing histories.
• resource_pool - Allocate a pool of shared resources across workflows.
• schedules - Demonstrates a Workflow Execution that occurs according to a schedule.
• sentry - Report errors to Sentry.
• sleep_for_days - A workflow that runs forever, sending an email every 30 days.
• strands_plugin - Run Strands Agents as durable Temporal workflows (model calls, tools, MCP, HITL).
• trio_async - Use asyncio Temporal in Trio-based environments.
• updatable_timer - A timer that can be updated while sleeping.
• worker_multiprocessing - Leverage Python multiprocessing to parallelize workflow tasks and other CPU bound operations by running multiple workers.
• worker_specific_task_queues - Use unique task queues to ensure activities run on specific workers.
• worker_versioning - Use the Worker Versioning feature to more easily version your workflows & other code.
• workflow_streams - Workflow-hosted durable event stream via temporalio.contrib.workflow_streams. Experimental

Test

To run the tests:

uv run poe test