How do you use LangSmith to inspect the exact message array sent to the LLM at every step of an agent run?

Q: How do you use LangSmith to inspect the exact message array sent to the LLM at every step of an agent run?

When you enable LangSmith tracing (set `LANGSMITH_TRACING_V2=true`), every invocation of an LLM (including those inside agents) creates a trace. In the LangSmith web app, you can select a trace, expand the LLM call, and see the exact `inputs` object which contains the `messages` array that was sent to the provider. This includes all the role/content pairs and tool call definitions. You can also attach custom tags or metadata to runs to filter them. For programmatic access, you can use the `langsmith` Client to fetch runs by ID and examine the inputs. ```python from langsmith import Client client = Client() # Get the latest run for a given project run = client.list_runs(project_name="my-agent-project", execution_order=1).first() # Access the inputs inputs = run.inputs # For an LLM run, inputs['messages'] contains the message list messages = inputs.get('messages', []) for msg in messages: print(f"{msg['type']}: {msg['content']}") # You can also query by trace ID trace_id = "your-trace-id" trace = client.read_run(trace_id) print(trace.inputs) ``` In the LangSmith UI, the trace view also shows token usage, latency, and any errors. This is invaluable for debugging why an agent made a specific tool call or to verify that the conversation history is correctly constructed.

Difficulty: 6/10

LangSmith tracing, message inspection, agent debugging

LangSmith automatically traces every LLM call within an agent run. You can inspect the exact input messages by viewing the trace in the LangSmith UI, where each LLM run shows the full `messages` array including system, human, AI, and tool messages. You can also programmatically fetch the trace via the LangSmith SDK.

When you enable LangSmith tracing (set LANGSMITH_TRACING_V2=true), every invocation of an LLM (including those inside agents) creates a trace. In the LangSmith web app, you can select a trace, expand the LLM call, and see the exact inputs object which contains the messages array that was sent to the provider. This includes all the role/content pairs and tool call definitions. You can also attach custom tags or metadata to runs to filter them. For programmatic access, you can use the langsmith Client to fetch runs by ID and examine the inputs.

How do you use LangSmith to inspect the exact message array sent to the LLM at every step of an agent run?

Difficulty: 6/10

LangSmith tracing, message inspection, agent debugging

LangSmith automatically traces every LLM call within an agent run. You can inspect the exact input messages by viewing the trace in the LangSmith UI, where each LLM run shows the full `messages` array including system, human, AI, and tool messages. You can also programmatically fetch the trace via the LangSmith SDK.

How do you use LangSmith to inspect the exact message array sent to the LLM at every step of an agent run?

How do you use LangSmith to inspect the exact message array sent to the LLM at every step of an agent run?

Scenario Questions

Follow-up Questions