Skip to content

README.md

Latest commit

 

History

History
38 lines (26 loc) · 1.59 KB

README.md

File metadata and controls

38 lines (26 loc) · 1.59 KB

LangSmith Tracing (Graph API)

Demonstrates combining the LangGraph plugin (durable execution) with Temporal's LangSmith plugin (observability) for full tracing of LLM calls through Temporal workflows.

What This Sample Demonstrates

  • Using LangSmithPlugin on the Temporal client for automatic trace propagation
  • Using LangGraphPlugin on the Worker for durable LangGraph execution
  • @traceable in three places: on the Activity itself, on a helper called from inside the Activity, and on a helper called from inside the Workflow
  • Both plugins working together: durability + observability

How It Works

  1. The Temporal client is created with LangSmithPlugin(add_temporal_runs=True).
  2. A Worker is created with LangGraphPlugin wrapping the chat graph.
  3. When the Workflow runs, the chat node executes as a Temporal Activity.
  4. @traceable decorators emit trace data to LangSmith for the Activity, an in-Activity helper, and an in-Workflow helper.
  5. The LangSmithPlugin adds Temporal-specific metadata to the traces.

Running the Sample

Prerequisites: uv sync --group langgraph and a running Temporal dev server (temporal server start-dev).

export ANTHROPIC_API_KEY='your-key'
export LANGCHAIN_API_KEY='your-key'

uv run langgraph_plugin/graph_api/langsmith_tracing/main.py

Traces will appear in your LangSmith dashboard.

Files

File Description
workflow.py @traceable chat node + helpers, graph definition, and ChatWorkflow
main.py Starts a Worker and executes the Workflow in a single process